Back when I was a laser pointer enthusiast, I was in a forum where several of their members collected ion lasers. These things need to be turned on regularly or else they die. Not so bad for argon ions, but HeCd lasers required about 24h turn on time distributed along each week to not die, and some were even worse. There was a guy who literally heated his house with the lasers, and he had about 20 or 30 ion lasers. Poor guy got broken in and the burglars broke most of them and had to sell the remaining ones.
@@douro20 not sure about copper vapour but for sure on argon and ion. Worse is the mixed gas.. Argon is 0.1% efficient... Argon and krypton mix is 0.01% efficient.. And more sensitive to gas pressure changes that change as you ran your system
Also Dave to check if your tube is still good: use a neon tube tester and check for a glow.. You can get those cheap on Amazon or ebay (the ones with that weird antenna on it). Be careful the tube holds a deadly charge and it's very fragile but very heat resistant quartz glass..required seriously skilled workers to make these that are rare to find these days.. That plasma was so hot that if you didn't cool it it would in theory would try and get hotter then the sun and fail ofc
I can answer any question if you want.. However if you want to talk to a guy who still refurbishes and sells tubes then you should talk to Brian Bohan of lexel laser..one of the only guys left as far as I'm aware.. The amount of work in pumping this tube under vacuum (roughing pump and molecular pump and heating it immensely) is beyond insanity and the gas purity had to be somewhere in the levels of 99.999999% or something
I'm an engineer at Coherent. Most of the heat (about 80%, depending on mains voltage and resultant B+ rectified levels and tube current setting) is dissipated from the tube into the cooling water flowing over it. The solenoid tube magnet and the PSU linear passbank dissipate the rest. The brass rectifier diode mounting block contains an impeller wheel that tracks water flow rates to shut down the laser at a defined trip point. There are many proprietary design details involved that prevent the ceramic plasma tube cracking from the heat dissipation along the tube. The white dust on the chassis is the result of the PSU having been subjected to condensation that corroded the chassis zinc plating, probably from the environmental conditions related from mobile use in light shows. The damaged varistor probably suffered from condensation dripping on it and shorting out. I think I see white water staining residue on the PCB at that location. The passbank has a plastic drip tray to contain condensation, but sometimes condensation forms on the PCBA in abusive environmental conditions.
On the brass heat dissipation block, are the two smaller four terminal devices actually bridge diodes? I have some very large MOSFETs from a neutral charging board that came out of a MGE UPS that look nearly the same..
@@colt4547 I can't get into that level of detail on those exact devices as proprietary information because these laser systems are still being manufactured. In spite of the low wall power conversion efficiency compared to solid state lasers, the Ion lasers still enjoy a healthy market due to the superior beam quality they provide.
Hey Dave. Lasershow specialist here. I do laser light shows for a living and also have an ion and the same power supply. No you cannot modulate a ion laser... You use something called a pcaom which is like a quartz crystal where you pump in frequencies to pick out a wavelength range (color) from the beam to then feed to your galvonometers. This was your modulation and it's extremely linear and accurate and fast! You had tunes like the coherent Sabre that did 30watts! And mixed gas which did full color from one beam (argon and crypto gas mix). The purity and cleanlyness of the tube is like a neon tube on steroids... The gasses needed to be very pure and the inside of the tube insanely clean.. The problems you talk about is that the pressure can drop or micro cracks could form on the tube with age causing all kinds of problems. They don't use these things anymore for laser display because of their enormous power requirements and poor efficiency and requirement of high flow water cooling. There is a solid state technology that matches/beats it which is opsl and vecsel.. Both are very expensive technologies however
I've seen ceramic tube resistors like that, but much bigger. They were about a meter long, with about 4 or 5 cm inside diameter. They were used as loads during the early testing of the electric power system for the ISS. The engineer who built that load wanted to water-cool them simply by running water through them, with 120 Vdc across them. The resistor manufacturer said OK, but the MUST orient them vertically to avoid air bubbles. So of course the guy mounted them horizontally, and they cracked, spewing water all over the place while the high voltage was on. After that, management decided NO MORE WATER-COOLED LOADS! So when we built this huge lab full of big loads, there was this giant air conditioner on the roof. Working inside the lab you had to put your jacket on whenever the AC was on, and take it off the rest of the time, on about a 20 minute cycle. Eventually the AC unit caught on fire and there SO many fire trucks.
Forgot to mention, the AC fire on the roof was such a huge panic because the AC was so heavy they feared the roof might collapse, crushing a whole floor full of custom-built electric test gear, and maybe even crash through that floor to the clean room below where all the space hardware and our thermal-vacuum chambers were. Fortunately, the roof held. But not long after that the whole north wall detached from the building and was tilted; only a few pipe kept it from falling flat onto a tank farm. This was caused by the Northridge earthquake in 1994. The sprinklers turned on and the water turned all the acoustic ceiling tiles into mud, and they collapsed onto all our test equipment, getting all over and even inside the cabinets.
@@steverobbins4872 Since then, the resistor manufactured modified the datasheet : "Warning, improper water cooling may cause earthquake and building collapse". That is some seriously intricate course of events.
I once pulled apart a 50W output CO2 cutting laser, made in 1983. It used two valves for current regulation, none of this transistor rubbish! They're Eimac 450TLs, which I still have. These pre-WWII designed valves have an anode dissipation of 450W each!
We used pair of its big brother (30kW in/20W out) to power the prototype for LIGO in the late 80's. Same type of passbank but 3x each of those clusters, over 100 transistors. P/S was like a small refrigerator in size. Voltage had to be higher too as the discharge tube (laser head) was at least twice as long. I hated working on that thing, it failed constantly. A beautiful day when LIGO switched to solid state Nd:YAG lasers in the mid-90's
This is a good example of a magnetic amplifier (magamp). The collector current biases the transformer core to vary its saturation current. This changes the amplitude of the secondary voltage, which would look like a square wave. Also, I think the reason for using bipolar pass transistors instead of MOSFETs is for loop stability. Bipolars have the disadvantage of requiring significant base current, but their advantage is a lot smaller capacitance from C to B. I've never userd a plasma tube before, but if it's an inductive load, the feedback loop will be more stable with bipolars. A good experiment for beginners: do a SPICE simulation of a simple current regulator with an inductive load, and see how the open-loop Bode plot changes as you go from MOSFET to BJT. It's quite an eye-opener.
Are you sure about the magamp? The secondary of the transformer would have DC current through both legs, so the DC current cancels out and does not contribute to saturation of the cathode transformer.
@@kubeek Damn, I think you're right. Thanks Kubeek. And sorry to everyone else. So how does this thing work? If it's not a magamp then they basically are just running the heating coil from the mains unregulated, so the heat would change as the mains voltage varies? Seems like poor design. Also, the pass transistors are just regulating the current through the plasma tube, but what produces the dc voltage between the heating coil and plate? The "magnet" coil seems like it is just to keep the beam of ions focused.
It's not just an inductive load, the impedance of a plasma tube is essentially *negative*, meaning that more voltage (like a spike from an inductor) will result in proportionally more current and lower resistance. So having a damping element in your control loop is essential, because any fast switching elements would result in a runaway and serious damage.
@@steverobbins4872 The pass transistors are connected to the cathode of the tube (on the center tap), while the anode is just at fixed voltage (DC positive).
@@Stoney3K Okay, but the collector of the BJT just connects to the xfrm center-tap, and thence to the tube's heating coil (the anode?); there are no other connections. So where does the collector bias voltage come from? It seems that the tube somehow creates the voltage, but I don't see how. Wikipedia doesn't have anything on "plasma tubes".
A bit of trivia Coherent originally called itself Coherent Radiation but the US postal service and other shipping companies disliked shipping packages and crates with word radiation printed on them so they shortened the name to just Coherent.
Cor Crikey, Look at this little beauty a 1 Farrad Capacitor in it's natural environment I'm just going to put the case back on this bad boy because i don't want to upset the old girl and i understand that i'm interferring with her natural environment and she's only upset because she's protecting her younger capacitors that's ok girl, i'll put the cover back on and make sure to replace all the screws and then i'm going to slowly back again SHE'LL BE RIGHT LOL
Steve tended to be irritating to most Australians Dave doesn’t have that effect on me. (Steve did some good things for conservation in Australia mostly by galvanising others and after his death demonstrating how a corrupt government can screw the environment and citizens)
@@glenecollins That's not even remotely true How do you figure ??? Steve, AS AGREED UPON BY THE MAJORITY OF AUSTRALIANS was a top fucking bloke. and Non irritating to the majority Maybe there is a person here and there that found him irritating, sure, i can understand that but mate, THE MAJORITY, BY FAR found him entertaining but yeah.. we all love dave
@@martinkuliza he was too “over enthusiastic” to keep a place on Australian TV at least until he was really famous in the US. That said I doubt there are too many Australians who don’t think he was a great bloke (except for the buisness people and politicians him and his foundation have prevented making a quick buck ripping up more of QLD.) He had a big heart and a lot of push
Ya got it's sort of right but close enough 4geeks speak! But I digress. I actually took a screenshot of when the first parts of the video showing the size and the plug of the power supply. I had and still have the old RadioShack power supply not the Wimpy little unregulated Black Box it looks more like a project box or switch Terminals circuit breaker in line cord. But one that looks like a proper power supply chassis what looks like fins on the outside. In text I sent with screenshot saying it remember that power supply insane this is not a power supply this is a power supply in explaining just what this was for how much power Etc! Not sure but I'm pretty sure it Rose eyebrows at the other end. Probably even if you were to try to hook it up where I would be even if you did have a proper plug you probably still might knock out block. Even though there is 3 phase in the area! If you had access to all three phases and a proper connector that is! You might be able to get the two of the three but good luck. Probably one phase would be in excess of capacity of incoming Supply. Even if it were to 240 AC. 120/ 240 here. Possibly 208 volts ac. Not sure if where I am is is tapped off 3-phase like previously. I do know they are for sure I had 208 where one would expect 220 or 240! Plus hundred percent Electric which Wayans decrease the efficiency for anything using 220-240 . However 120 was within normal spec. So no worries there but still not as efficient as could be lift only straight up split phase power instead of cap from two of three phases! Just something to be aware of if you're in that situation! Which also would translate to higher electric costs due to inefficiency because of lower voltage on 220 240 applications
Thanks Dave, brought back so many memories. In a former company I used these Ion lasers (Krypton, 413nm) for Photoresist exposure of Glass masters for creating CD/DVD metal masters. We serviced and replaced the Ion tubes like every 6-8K hours of operation. Certified thru Coherent to purchase tubes, troubleshoot and replace in house. Replacement cost for Kr tube and mirrors was like $24K. I have all the service manuals and schematics on a CD-ROM. I also repaired several of the power supplies, very interesting design. The tube and PSU require 18KW of cooling capacity and ~16KW of 3 phase 208v power. Very heavy and so well built!
Is possible to introduce more argon inside the tube to still lase? I have a no lasing INNOVA 308C and I want to introduce some Argon into the tube than to vacuum it.... It will work?
Nice to see KANTHAL mentioned on the EEVblog! They specialise in heating elements and products such as the one you found. Comes from my hometown of Hallstahammar. ~ 25km from where ABB was founded.
The name is also used for a group of chemical resistant alloys used in resistors and heating elements. "E-cigs" use these alloys in their heating coils.
Kanthal is a resistive wire material. So the company got its name from the material, or they developed that alloy and simply their name is stuck to it, as evebody knows it now as Kanthal wire. Like as we call those mini high RPM cutting/grinding machnies a Dremel, although not only Dremel makes them.
@@mrnmrn1 the founder Hans Von Kantzow was a mountain engineer and while working in Degerfors 1916, he forgot some test rods in a furnace and observed that the steel could handle the high temperature. This lead him to later file a patent 1926 after he had studied the test rods further.
The "big industrial circuit breaker thing" looks like a contactor for switching the high current path and the power for the contactor coil goes through the thermal fuse to shut off the rectifier if it's overheating
Either that, or current limiting for the main power capacitors on startup. The contactor closes after about a second or so after power up, and while the contacts are open there is a big current limiting resistor in series. At the end of the short period, the contactor closes and shorts out the resistor. Critical to managing the main electrolytic caps' longevity. This is very common in 3 phase Variable frequency drives as well for the same reason. I measly NTC wouldn't cut :)
@@nohandle1 Not really the SCR pre regulator in there will do a pretty good job of slow start, as it will be driven from some form of controller, likely a dedicated Telefunken chip on the lower board, so that it has soft start using a capacitor, and then regulates the voltage on those capacitors.
@@SeanBZA ah yep, that's does make more sense. I did wonder what that was in there for. In VFD's sometimes a single SCR is used for regenerative/dynamic breaking (3phase motors), but that's a totally different application.
Another use for a three-phase contactor on the front end is for rectifier protection. If phase monitoring determines a phase was lost on the input, the contactor opens to prevent over-current in the rectifier. This is common in large UPS systems.
Right, make no sense build in an unreachable breaker, when you had fuses they are reachable from the outer side of this case. The correct name for this "big power relay" is contactor.
Nice teardown Dave, that passbank would make a nice showpiece, I was a bit surprised at the low transistor count, but it all adds up. I was curious as to where the igniter was hiding, and it makes sense that it will be in the head itself as it needs to generate around 10kV to start the arc.
Greetings from Germany to the Neighbors:) Schrack is a German top quality industrial relays manufacturer: www.schrack-technik.de/kontakt I didn't know they produce in your country too. The world is small:)
Liquid cooled Argon lasers setup at 90's Raves where amazing and cost mega cash to hire . The laser head would have a seperate box of prisms and mirrors to make the FX . I can't believe this was getting ditched.
Most likely the tube was at the end of its lifetime, and a new semiconductor laser was cheaper then a replacement tube. Which costs tens of thousands of dollars.
most argon krypton or argon lasers are being ditched. and are being sold for pennies or are given away for free. They serve little use because of their requirements and size.. now people want more projectors which even in terms of power and water requirement is just not feasable when going even in numbers of 4 units..
@@Masterpj555 Interestingly enough, those old ion heads had very elaborate optical decks to generate effects which really can't be matched on simple semiconductor lasers with deflection galvos. They basically had to do the same 'wow' factor you get out of a dozen projectors, from a single head and optical setup -- fiber launched split setups did exist but they were not common (and cumbersome to set up). Pink Floyd was infamous for having very impressive laser displays, which were all controlled with analog multi-track tape players and even synthesizers.
@@Stoney3K wow ! had no idea how Floyd controlled the lasers but now it makes sense ! remember the lighting tech/roadie lighting his cigarette off the beam at the start of the delicate sound of thunder concert ?
Dave, I was a physics research student and was lucky enough to run my research project using a brand new Coherent Micra oscillator and legend elite amplifier titanium sapphire set. Just a few weeks before my thesis defense I was running some tuning and managed to accidentally miswire a high voltage supply into the fast photodiode detector port on the controller board. Needless to say, I let the magic smoke out and delayed my graduation by a few months whilst I got it repaired. I love the channel, keep up the great work!
Hi Dave! In modern labs you won't find Argon Ion lasers anymore. DPSS are commonly used these days, like Coherent Verdi. Better efficiency, single phase powered with the same beam quality.
@@sudazima Microscopy lab? That is certainly an application where it is still sometimes used, although even there they are slowly replaced. There is a reason why so many fluorphores are 488nm
For more demanding outdoor applications like Laservision is doing, DPSS isn't used, instead, Coherent's OPSL is the tool of choice. Even though the beam quality may be *slightly* less than an ion, the logistics of just plugging in a unit on the mains and rigging it like a regular lighting fixture really outweighs the hassle of running all the infrastructure for an ion head.
Ran a few of Coherent and large frame Spectra Physics for entertainment in the mid 90ies. The heaviest part to unload were the transformers to bring european 400V three phase to american 208V for midframe ion lasers or 480V for the large frame SP 171's. We always tapped the transformers so that the voltage across the passbank was just high enough to be in regulation, as not to stress the power supplies. We were using mostly used ion laser heads from universities that did not meet their specs for the scientific work anymore, but that meant that these heads tried to kill the power supplies by having low voltages across the tube and running at higher currents for acceptable output. So we were used to blown passbank transistors and repaired many exciters on the road, just before a show was about to start. The older generation power supplies without microcontrollers were much better in that regard, with a real knob for current, adjustable magnet current, manual ignitor with unlimited retries to strike the tube and analog meters on the front.
You know what the pass transistor array needed? More warnings about high voltage. I don't know about you but when I see huge traces, lots of component isolation, and monster AWG cables sticking out of it, I just want to touch everything. Only having nine warnings is playing with fire.
I use a bigger, better one (a "Coherent Innova Sabre") at work quite regulary. It can do multiple wavelenghts and can be used both in multi-line and single-line operation. Tube runs at around 500V at up to 55A, so 27.5kW into the tube, around 40-60kW from the mains. Output in single-line operation depending on the chosen wavelength around 2-12W. To make it more fun, we use it as a pump light source for a dye-jet laser, which itself has somewhere around 3-20% efficiency. For your information: "LIGHT" switches from Constant Current operation to regulation of optical (light) power. At power-up you typically raise the current until you are roughly at the desired power level, then switch over to "LIGHT" and set it where you want it. "PWR TRK" is "Power Tracking". It's a set of actuators additional to the vernier screws for the back mirror, which does continuous fine tuning for maximum power. This usually is an option you have to pay extra for, but the button is always there. The problem with those lasers is, they use up their gas during operation. To compensate for this, the tube has a reservoir and a valve to refill when necessary. The problem is, the valve isn't perfectly sealed, so some of the gas is diffusing into the tube all the time. If you leave them powered off for say 2-3 years, the pressure will be too high for the igniter to ionize the tube and you will need a new one.
Ahh, the good ol' days. I can imagine the young blokes in the workshop standing over the power supplies soldering and routing and adjusting the metal work on a grinder, screwing and testing - days of assembly work from beginning to end. Dave, I hope you tear down the laser too.
Hey thought you'd show up here somewhere! Also would have been nice to even just that which those big old Power resistors you know the aluminum case ones. I have heard on some of those if you put enough current threw them under right conditions they go off like a shotgun shell. I've heard people doing that deliberately girl scared the bejeebers out of someone. Or like those other times where you stick some sort of power resistor or something across the power on a big power supply unbenounced to someone else. Anything that would let out smoke a nice bag Etc. Never done it myself but I've heard about people doing it. Someone told me personally today at dumb it to an instructor setup. Between classes before his class don't know how he got in there and accomplished it but he did and. Let's just say dad instructor. Which I meant later said that was probably the most memorable set up he ever had. At least when the biggest things that happened to him personally in his teaching career. You remember when somebody goes off like a shotgun shell that you don't expect to have that happen. Sort of like when a bag of filter caps all go off at once due to some sort of failure in a power supply. Only had that happen once but remembered definitely remembered. Bad shutdown circuit over-voltage by at least double or more! All it was was one bad resistor took out the shutdown circuit no Fail-Safe
I used for work at Spectra Physics as an engineer working on DPSS picosecond/femtosecond lasers, and in college, I used to work as a repair technician on mostly ion lasers from both Spectra and Coherent. One thing of interest is the concept of power mode and current mode. I'm sure "light mode" puts the tube in closed optical loop operation where an attenuated and properly diffused optical signal is referenced by a photodiode to put the tube in close loop control to maintain a stable light output level. I had a lot of fun applying (discrete-time) control theory to lasers. Using microcontrollers affords a lot of flexibility. At the big convention centers, the sudden inrush of thousands of people on the floor changed the local environment that we could see it in the data-loggers for the control actuation signals, and we were pretty sure that a lot of other vendors were affected as well. We even used water-cooled power meters to verify the power stability. One thing you have to be careful with with closed loop operation is guaranteeing that the beam pointing does not wander, which becomes harder when there are more optics in the box. It was solved for the laser I worked on, and am curious about how this is handled in general.
Those gas heads were the cream of the crop for laser displays for years, in particular the Argon-Krypton mixed gas variant which actually produced a white beam, which could be modulated with a PCAOM to select colour. I used to tinker around with a 1980s Spectra-Physics model 168 which was a few watts of white lights out, and about 15kW of electrical power in. All of the excess heat was dumped into the cooling water, not just from the head, but also from the (linear) power supply. The regulation in the PSU was still done by a tube! I also had an ILT 5940A ion laser head which was air cooled on my hobby bench.
I absolutely love the operator warning on the side of the laser. I need to get some of those stickers. Anyway, repair it and turn it on! I bought a HeNe gas laser probably 25 or 30 years ago because I wanted to get into making holograms, but due to procrastination, lack of money, and lack of a good place to set it all up, I've never actually started it. So, the laser is mounted on my telescope for -nefarious purposes- ...I mean, for staking flags on my property lines between pairs of survey markers.
I get giddy when you breakdown stuff from the entertainment industry. when times were better I work in sound and light for entertainment purposes. As times goes I have been learning so much about EE in general. you rock, Dave!!!!!
At 29:48 that's not a circuit breaker, but a contactor (big relay with current break capacity). Software probably starts up and completes power on self checks before pulling in the contactor to apply power to the output stage.
Daryl is good people! He and I spoke and I thoroughly enjoyed the podcast. Sad to see that amazing laser torn down :( If you ever need a person to interview on laser tech for a Podcast do find me!
The plasma tube is filled with about 350 Millitorr of Argon. The arc has a negative impedance characteristic that makes designing the control loop very interesting. The tube will take all the current you can feed it, and will try to "run away" if the current is not limited. I've rebuilt the tubss and power supplies for those and Coherents competitors for about 30 years now. You'll find the hot cathode in the plasma tube needs around 28 amps at 3.3 volts give or take. The secondary of the cathode transformer is center tapped, and the tube current flows off the center tap onto the high side of the passbank. The transistors almost always fail shorted, so there is a diode and fuse in each base lead and typically a fuse in the collector on the 36 transistors. The 300 series is super sophisticated, an eeprom is created at the factory with the i-v characteristics of the plasma tube, and if the tube doesn't match the eeprom, it warns you for the need for a gas fill or shuts down. The plasma tube you have does not contain Beryllium Oxide, but the insulators under the To3s almost certainly does. I have the field service manual, but I can't post it.
The one thing you're missing is diode lasers are always switching you cannot run them 100% duty cycle (then) 90s. 1950s lasers still existed to make light shows look good on ccd camera. With polarizers for color and mechanical mirrors... And by the way you meant collimated. Luv ya dave!
I had a very small lasershow business in the 2007-2015 period, that was the time were high power laser diodes started to appear. I saw the business of these things basically being destroyed in few years and now you can buy a 10W green laser projector (the whole projector, not only the laser) that is half the size of that power supply for 1/10 of the price of that laser module. One club where I worked still has one of these beasts. It was still in weekly use when I left in 2015, I would love to go back there and check it out now 🥲
This takes me back I was a physics postgraduate student and spent two years with a similar unit the Innova 300, think max power out was 30w but we never got more than 24 out of it. My first job was to plumb the water cooling but opening it up by myself was definitely a no no! One of my supervisors had to tune it when it was feeling a bit crook. Total beastie!
The fuses on the passbank are there to prevent a fire. There's a huge amount of energy flowing through such a tiny box. I remember swearing at the designers as a technician in college, only to go to lunch with some of them when I joined as an engineer. Ion laser supplies often fail pretty spectacularly, especially the early designs .. Tubes going up to air for example, created a complex impedance that caused a reflection back to the power supply. Please note that the passbank has one set of transistors for the tube, and one for the solenoid magnet to focus the beam down.
I feel like the "disadvantage" of the overall current dropping if one transistor blows could actually an amazing advantage, because the monitoring could basically divide total current by the requested per transistor current to get live feedback of how many transistors are actually working, and therefore how many have blown. "Dear maintenance staff, 2 of my transistors have blown, I can still compensate and operate just fine, but please replace." Of course, this is all speculation.
@@TheHuesSciTech Well .. I was a little fast here .. you actually need to measure individual current, too .. but since they are all tied to a common potential, a simple analog multiplexer would be enough for this.
The large white block inside, next to the caps looks like a 3ph contactor. Probably bought in after the caps have charged through a resistor. Maybe that weird looking one on the underside you removed? A lot of large motor inverters use this method also. I have some broken IGBT modules I keep meaning to send in ~1.7kv 450A, you can remove the top and see the silicon components/bond wires through the gel. I think the max thermal dissipation is around 2KW per package.
LOOK AT THAT BAD BOY. 97??? Fantastic Vintage😁 Dave??...Your A Highly motivated Peeps generally! So, meeting Peeps like Darryl, does it give you a spring in your step? And keep things running positive? Cos even if a Peep has everything , they still could feel, like it's same old, same old?? I just find it fascinating how you've moulded your lifestyle to work for you. When most of us feel like we don't hold all the cards. Well done.
The SCRs are probably for crowbar protection against overvoltage. If the voltage goes over the critical limit, the SCRs turn on (and latch), grounding the outputs, which blows fuses / trips breakers. Might do a little damage in the PS, but not too bad, (something needed to be repaired anyway!) and saves the laser. Was just talking with someone today that told me he lost a radio recently due to his PS shorting a pass transistor, which then dumped 32v into his 12v gear. Wouldn't have killed his radio if the PS had included a crowbar. I know all three of my big PS have crowbars.
I still use one like this, but slightly older Zeiss. One thing about this type of laser - there is absolutely no way to shift the lines and they are extremely narrow. When I need to double check my equipment alignment there is nothing to replace our trusty old argon laser. And due to it's purity it's easy to filter it off, so it also works well as a source of excitation for some of my materials.
At 22:17 it looks like one of those transistors may have blown up too. On the row closest to the board edge, four away from the backplane connector. That was a fun teardown. Astounding inefficiency in that system. Now I need to see if you tore down the head too. It'd be neat to see a it powered by a neon power supply (if it'd even work a little) to see that massive tube glow just a bit.
I remember an old Unholtz-Dickie shaker table amplifier, rated something like 10kW -- poor thing stood three or four racks across, huge 3ph transformer in the bottom of one, bunch of fancy calibrated circuitry in the control unit (for displacement, acceleration, etc. control, sweeping, etc.). The power output I believe used a Circlotron architecture (tube aficionados know about that one!), and consisted of a Metric Shitload of stud type NPN transistors on a common rail, for each output. The rail being a water cooled tube.
Have a few Coherent's in my lab. A 1W CW, 500mW CW OPSL and a 2.5W Q-Switcher, pricey kit but really good for the money. Gonna echo @Filipamator 's comment about how most lasers are DPSS or OPSL technology these days. Also currently working on getting a 20W Verdi. M2 on the CW lasers are 1.02 on the 1W and 1.05 on the 500mW unit, so solid state can definitely hang / surpass these days.
I'm loving the "I'm pretty confidant I can do..." as you oh so delicately tap the caps, lol. Be honest, you weren't that confidant (and to be honest I wouldn't either; you never know if a resister has failed)!
I'd be very very confidant. Of course, behind the scenes I'd have clipped a 470k ohm resistor across the capacitor leads half an hour earlier, and only removed that resistor maybe a minute before I started filming.
It looks like those fuses on standoffs would not be fitted initially. The standoffs would be used as the measuring point for the current. The fuse would replace the ammeter once current was set for each transistor.
Used the larger Version of that with up to 25 Watt of multiline output to pump a coherrent 899-29 single mode dye laser and a Spectra Physics 171 on the same table for a second 899... Had to build an own heat rxchanger for the old stuff, the Coherrent Sabre had its own. The dye lasers were used for spectroscopy of liquid Helium cooled samples at 1.5K. We got a New 100kW power line for that Lab ;-)
The white industrial block thing is not a circuit braker Dave! It is a huge contactor/relay. They brobbably can kill mains power electronically if they need to, or ot is there as a precharge stage together with a resistor to precharge the huge smoothing caps to limit the inrush current.
I've dealt with several permanent laser shows around the US. we had a pair of Coherent 80w krypton/argon white light lasers. Their product is second to none!
Love it 😎 I have a few of these. Spectra physics 168 Argon Krypton mix. A 2020 and a coherent sabre.. it's a beast 30watts. 208v or 480v 3phase input on these so you need a giant 3 phase transformer aswell.. lol fun times
I believe they would had used these for the blue beam, a semiconductor laser for the red beam and a copper vapor laser for the green beam. I didn't know this until recently but they actually make even more powerful argon lasers than this.
Yep, Mike's electric stuff is right. A Krypton/Argon mix will do multiple wavelengths across the spectrum from red to violet, so-called "white-light" ion Lasers.
Maybe the rows and column transistor array is supposed to allow like for 1 or 2 blown transistors and still work. Like if a column lost one they could have been running underrated and just crank up to the same level using the remaining ones.
These things would operate in current mode, as you describe, but also in "light mode", where an optical pick-off in the head would feed back light output to the psu (using a pin photodiode) to give a very stable output, hence a lot of the extra complexity I guess (ie it is constant current AND variable current) The water cooling cooled the passbank, then went out to the head, came back to cool the rectifiers, then exited via the "drain" , normally to the input of a heat exchanger, where it would be cooled before being sent back to the passbank input. Water bills get high fast if you use the drain as an actual drain ;)
In the well-funded lab where I did my PhD, we used fiber lasers with harmonic generators to replace this sort of laser. Much more efficient. Each one cost about $70000 though, and you needed three because they were not really tunable. They did have very good beam quality, as they were in the end solid-state lasers not diode lasers.
That white dust caked on is likely zinc galvanized metal oxidizing over time to zinc oxide. Benign as long as you don't heat the metal up to welding temperatures or try snorting it....
Are you sure the 4 screw devices you identified as bridge rectifiers are actually bridges? I used similarly packaged darlingtons in the past. They had great specs and current handling and they were used in dc motor control devices.
I'd take a guess and say that they could be SCR's in a controlled 3-phase rectifier arrangement, much like the type you would find on variable frequency drives.
@@Stoney3K If that’s the case shouldn’t it have 3 devices instead of 2? That might be a good guess though, those packages puzzle me, I’d love to see what they are! @eevblog any chances to get some pics of the part number?
I'd take the EDLC off the mainboard and save it for another project. The reason for the high amount of insulation isn't for the operating voltage, which is around 260 volts, but rather the ignition voltage which is around ten kV.
i found a Coherent Eye Surgery Medical Laser in the scrapyard a few years back. Best $50 i ever spent. It had a nice blue/green argon laser. Also heavy duty block diodes and mosfets mounted on a liquid cooled block. The red HeNe aiming laser was smashed unfortunately. The interweb told me to be careful with the ceramic laser tube and not to break it or make any dust because its toxic Beryllium Oxide !!!! The lenses and mirrors are possibly sapphire ???? Cha Ching $$$$
Dave I bought a ionization UV machine. I think it was a knock off from China? It has a decent circuit board on it with two high voltage Transformers. It's called "New Age living C 3500 high-efficiency air purifier" You definitely do not stay in the room with this thing on! It was covered with what looks like the white salty substance. I found out while cleaning, it water soluble. Much easier to wash off than to scrape off!
Pftt ! I use to be part owner of a laser cutting service. Used a 4KW Trumpf CO2 laser. That's 4KW of actual light output. 10s of KWs going into the machine. Needed 440V 3 phase power, with 250A breakers. Could cut steel 3/4" (18mm) thick !
I had one of those puppies in my Lab (albeit the more powerfull version with 35W all lines and 14W for one line). But i replaced it also with a diod pumped disc laser (with two times 20W laser power (one line) and less than 4kW electrical power usage. Those old Argon lasers are really power dumps ;). the power supply of our coherent argon laser was about 3 times the height and was impossible to lift even a bit (i could only roll it). and there was a second similar sized unit for the cooling (mind you that also needed an external cooling water supplied, just you need it very stable for the laser, that is why it neede the seperate cooling unit. I think power needed was 56kW? but i might remember it wrong :/. We were using it for an experiment in a synchrotron (with a peak power usage of about 3MW, so the laser was just one of the small fries ;) )
Back when I was a laser pointer enthusiast, I was in a forum where several of their members collected ion lasers. These things need to be turned on regularly or else they die. Not so bad for argon ions, but HeCd lasers required about 24h turn on time distributed along each week to not die, and some were even worse. There was a guy who literally heated his house with the lasers, and he had about 20 or 30 ion lasers. Poor guy got broken in and the burglars broke most of them and had to sell the remaining ones.
I didn't think vapour lasers were quite that sensitive. I would imagine copper vapour lasers aren't quite so sensitive.
@@douro20 not sure about copper vapour but for sure on argon and ion.
Worse is the mixed gas..
Argon is 0.1% efficient... Argon and krypton mix is 0.01% efficient.. And more sensitive to gas pressure changes that change as you ran your system
Also Dave to check if your tube is still good: use a neon tube tester and check for a glow.. You can get those cheap on Amazon or ebay (the ones with that weird antenna on it). Be careful the tube holds a deadly charge and it's very fragile but very heat resistant quartz glass..required seriously skilled workers to make these that are rare to find these days..
That plasma was so hot that if you didn't cool it it would in theory would try and get hotter then the sun and fail ofc
I can answer any question if you want.. However if you want to talk to a guy who still refurbishes and sells tubes then you should talk to Brian Bohan of lexel laser..one of the only guys left as far as I'm aware.. The amount of work in pumping this tube under vacuum (roughing pump and molecular pump and heating it immensely) is beyond insanity and the gas purity had to be somewhere in the levels of 99.999999% or something
Damn burglars! Level about the same chickens. Sh%t everywhere including their bed.
I'm an engineer at Coherent. Most of the heat (about 80%, depending on mains voltage and resultant B+ rectified levels and tube current setting) is dissipated from the tube into the cooling water flowing over it. The solenoid tube magnet and the PSU linear passbank dissipate the rest. The brass rectifier diode mounting block contains an impeller wheel that tracks water flow rates to shut down the laser at a defined trip point. There are many proprietary design details involved that prevent the ceramic plasma tube cracking from the heat dissipation along the tube. The white dust on the chassis is the result of the PSU having been subjected to condensation that corroded the chassis zinc plating, probably from the environmental conditions related from mobile use in light shows. The damaged varistor probably suffered from condensation dripping on it and shorting out. I think I see white water staining residue on the PCB at that location. The passbank has a plastic drip tray to contain condensation, but sometimes condensation forms on the PCBA in abusive environmental conditions.
that was great, thanks for the insight!
Thanks!
Engjneer Coherent
On the brass heat dissipation block, are the two smaller four terminal devices actually bridge diodes? I have some very large MOSFETs from a neutral charging board that came out of a MGE UPS that look nearly the same..
@@colt4547 I can't get into that level of detail on those exact devices as proprietary information because these laser systems are still being manufactured. In spite of the low wall power conversion efficiency compared to solid state lasers, the Ion lasers still enjoy a healthy market due to the superior beam quality they provide.
Hey Dave. Lasershow specialist here. I do laser light shows for a living and also have an ion and the same power supply.
No you cannot modulate a ion laser... You use something called a pcaom which is like a quartz crystal where you pump in frequencies to pick out a wavelength range (color) from the beam to then feed to your galvonometers. This was your modulation and it's extremely linear and accurate and fast! You had tunes like the coherent Sabre that did 30watts! And mixed gas which did full color from one beam (argon and crypto gas mix). The purity and cleanlyness of the tube is like a neon tube on steroids... The gasses needed to be very pure and the inside of the tube insanely clean..
The problems you talk about is that the pressure can drop or micro cracks could form on the tube with age causing all kinds of problems.
They don't use these things anymore for laser display because of their enormous power requirements and poor efficiency and requirement of high flow water cooling.
There is a solid state technology that matches/beats it which is opsl and vecsel.. Both are very expensive technologies however
About this pcaom..compare it to a variably controlled prism of sorts... But you could also control the intensity of how mucnh would be picked off
I've seen ceramic tube resistors like that, but much bigger. They were about a meter long, with about 4 or 5 cm inside diameter. They were used as loads during the early testing of the electric power system for the ISS.
The engineer who built that load wanted to water-cool them simply by running water through them, with 120 Vdc across them. The resistor manufacturer said OK, but the MUST orient them vertically to avoid air bubbles. So of course the guy mounted them horizontally, and they cracked, spewing water all over the place while the high voltage was on.
After that, management decided NO MORE WATER-COOLED LOADS! So when we built this huge lab full of big loads, there was this giant air conditioner on the roof. Working inside the lab you had to put your jacket on whenever the AC was on, and take it off the rest of the time, on about a 20 minute cycle. Eventually the AC unit caught on fire and there SO many fire trucks.
Looks like a fun and safe workplace.
Forgot to mention, the AC fire on the roof was such a huge panic because the AC was so heavy they feared the roof might collapse, crushing a whole floor full of custom-built electric test gear, and maybe even crash through that floor to the clean room below where all the space hardware and our thermal-vacuum chambers were.
Fortunately, the roof held. But not long after that the whole north wall detached from the building and was tilted; only a few pipe kept it from falling flat onto a tank farm. This was caused by the Northridge earthquake in 1994.
The sprinklers turned on and the water turned all the acoustic ceiling tiles into mud, and they collapsed onto all our test equipment, getting all over and even inside the cabinets.
@@steverobbins4872 Since then, the resistor manufactured modified the datasheet : "Warning, improper water cooling may cause earthquake and building collapse".
That is some seriously intricate course of events.
@@steverobbins4872 Geez. At what happened to the test eqipment? They got them repaired?
I once pulled apart a 50W output CO2 cutting laser, made in 1983. It used two valves for current regulation, none of this transistor rubbish! They're Eimac 450TLs, which I still have. These pre-WWII designed valves have an anode dissipation of 450W each!
450W each! Wow!
Could probably make a heck of a nice CW linear amplifier for AM long-wave ham radio with those tubes.
that is one of the famous water heaters with a coherent indicator lamp ;)
Nice one
We used pair of its big brother (30kW in/20W out) to power the prototype for LIGO in the late 80's. Same type of passbank but 3x each of those clusters, over 100 transistors. P/S was like a small refrigerator in size. Voltage had to be higher too as the discharge tube (laser head) was at least twice as long. I hated working on that thing, it failed constantly. A beautiful day when LIGO switched to solid state Nd:YAG lasers in the mid-90's
nice piece of history, thx
CD and DVD mastering was done with its little brother way into the naughties.
Fascinating! Go ligo!
This is a good example of a magnetic amplifier (magamp). The collector current biases the transformer core to vary its saturation current. This changes the amplitude of the secondary voltage, which would look like a square wave.
Also, I think the reason for using bipolar pass transistors instead of MOSFETs is for loop stability. Bipolars have the disadvantage of requiring significant base current, but their advantage is a lot smaller capacitance from C to B. I've never userd a plasma tube before, but if it's an inductive load, the feedback loop will be more stable with bipolars.
A good experiment for beginners: do a SPICE simulation of a simple current regulator with an inductive load, and see how the open-loop Bode plot changes as you go from MOSFET to BJT. It's quite an eye-opener.
Are you sure about the magamp? The secondary of the transformer would have DC current through both legs, so the DC current cancels out and does not contribute to saturation of the cathode transformer.
@@kubeek Damn, I think you're right. Thanks Kubeek. And sorry to everyone else.
So how does this thing work? If it's not a magamp then they basically are just running the heating coil from the mains unregulated, so the heat would change as the mains voltage varies? Seems like poor design.
Also, the pass transistors are just regulating the current through the plasma tube, but what produces the dc voltage between the heating coil and plate? The "magnet" coil seems like it is just to keep the beam of ions focused.
It's not just an inductive load, the impedance of a plasma tube is essentially *negative*, meaning that more voltage (like a spike from an inductor) will result in proportionally more current and lower resistance. So having a damping element in your control loop is essential, because any fast switching elements would result in a runaway and serious damage.
@@steverobbins4872 The pass transistors are connected to the cathode of the tube (on the center tap), while the anode is just at fixed voltage (DC positive).
@@Stoney3K Okay, but the collector of the BJT just connects to the xfrm center-tap, and thence to the tube's heating coil (the anode?); there are no other connections. So where does the collector bias voltage come from? It seems that the tube somehow creates the voltage, but I don't see how. Wikipedia doesn't have anything on "plasma tubes".
A bit of trivia Coherent originally called itself Coherent Radiation but the US postal service and other shipping companies disliked shipping packages and crates with word radiation printed on them so they shortened the name to just Coherent.
Your enthusiasm is infectious!! You’re like the Steve Irwin of electrical engineering!
Just don't send him any mechanical stingray for a teardown, OK?
Cor Crikey, Look at this little beauty
a 1 Farrad Capacitor in it's natural environment
I'm just going to put the case back on this bad boy because i don't want to upset
the old girl and i understand that i'm interferring with her natural environment
and she's only upset because she's protecting her younger capacitors
that's ok girl, i'll put the cover back on and make sure to replace all the screws
and then i'm going to slowly back again
SHE'LL BE RIGHT
LOL
Steve tended to be irritating to most Australians Dave doesn’t have that effect on me.
(Steve did some good things for conservation in Australia mostly by galvanising others and after his death demonstrating how a corrupt government can screw the environment and citizens)
@@glenecollins
That's not even remotely true
How do you figure ???
Steve, AS AGREED UPON BY THE MAJORITY OF AUSTRALIANS was a top fucking bloke.
and Non irritating to the majority
Maybe there is a person here and there that found him irritating, sure, i can understand that
but mate, THE MAJORITY, BY FAR found him entertaining
but yeah.. we all love dave
@@martinkuliza he was too “over enthusiastic” to keep a place on Australian TV at least until he was really famous in the US.
That said I doubt there are too many Australians who don’t think he was a great bloke (except for the buisness people and politicians him and his foundation have prevented making a quick buck ripping up more of QLD.)
He had a big heart and a lot of push
'That's not a power supply. THAT"S a power supply!"
Btw. the head would also be interesting to see.
Pawel Zadrozniak , Perhaps he didn't see the movie.
Ya got it's sort of right but close enough 4geeks speak!
But I digress.
I actually took a screenshot of when the first parts of the video showing the size and the plug of the power supply.
I had and still have the old RadioShack power supply not the Wimpy little unregulated Black Box it looks more like a project box or switch Terminals circuit breaker in line cord.
But one that looks like a proper power supply chassis what looks like fins on the outside.
In text I sent with screenshot saying it remember that power supply insane this is not a power supply this is a power supply in explaining just what this was for how much power Etc!
Not sure but I'm pretty sure it Rose eyebrows at the other end.
Probably even if you were to try to hook it up where I would be even if you did have a proper plug you probably still might knock out block.
Even though there is 3 phase in the area!
If you had access to all three phases and a proper connector that is!
You might be able to get the two of the three but good luck.
Probably one phase would be in excess of capacity of incoming Supply.
Even if it were to 240 AC.
120/ 240 here.
Possibly 208 volts ac.
Not sure if where I am is is tapped off 3-phase like previously.
I do know they are for sure I had 208 where one would expect 220 or 240!
Plus hundred percent Electric which Wayans decrease the efficiency for anything using 220-240 .
However 120 was within normal spec.
So no worries there but still not as efficient as could be lift only straight up split phase power instead of cap from two of three phases!
Just something to be aware of if you're in that situation!
Which also would translate to higher electric costs due to inefficiency because of lower voltage on 220 240 applications
The caked powder might be "white rust" or oxidized galvanized metal, fairly common occurrence with galvanized metals.
maybe it's alumina ( i.e. aluminium oxide or hydroxide) that formed as a result of corrosion due to an exposure to humid environment
The only other thing that comes to mind other than this would be smoke machine residue
Zinc oxide or aluminium oxide.
@@feosp Yes, I was suspecting Zinc Oxide myself.
@@feosp Correct. If the case is Zinc plated it will be definitely related to oxidation.
Thanks Dave, brought back so many memories. In a former company I used these Ion lasers (Krypton, 413nm) for Photoresist exposure of Glass masters for creating CD/DVD metal masters. We serviced and replaced the Ion tubes like every 6-8K hours of operation. Certified thru Coherent to purchase tubes, troubleshoot and replace in house. Replacement cost for Kr tube and mirrors was like $24K. I have all the service manuals and schematics on a CD-ROM. I also repaired several of the power supplies, very interesting design. The tube and PSU require 18KW of cooling capacity and ~16KW of 3 phase 208v power. Very heavy and so well built!
Is possible to introduce more argon inside the tube to still lase? I have a no lasing INNOVA 308C and I want to introduce some Argon into the tube than to vacuum it.... It will work?
Nice to see KANTHAL mentioned on the EEVblog! They specialise in heating elements and products such as the one you found. Comes from my hometown of Hallstahammar.
~ 25km from where ABB was founded.
The name is also used for a group of chemical resistant alloys used in resistors and heating elements. "E-cigs" use these alloys in their heating coils.
Kanthal is a resistive wire material. So the company got its name from the material, or they developed that alloy and simply their name is stuck to it, as evebody knows it now as Kanthal wire. Like as we call those mini high RPM cutting/grinding machnies a Dremel, although not only Dremel makes them.
Nice, I didn't know that. The first thought in my mind hearing that name was "nice cheese" :-)
@@mrnmrn1 the founder Hans Von Kantzow was a mountain engineer and while working in Degerfors 1916, he forgot some test rods in a furnace and observed that the steel could handle the high temperature. This lead him to later file a patent 1926 after he had studied the test rods further.
The "big industrial circuit breaker thing" looks like a contactor for switching the high current path and the power for the contactor coil goes through the thermal fuse to shut off the rectifier if it's overheating
Either that, or current limiting for the main power capacitors on startup. The contactor closes after about a second or so after power up, and while the contacts are open there is a big current limiting resistor in series. At the end of the short period, the contactor closes and shorts out the resistor. Critical to managing the main electrolytic caps' longevity. This is very common in 3 phase Variable frequency drives as well for the same reason. I measly NTC wouldn't cut :)
@@nohandle1 Not really the SCR pre regulator in there will do a pretty good job of slow start, as it will be driven from some form of controller, likely a dedicated Telefunken chip on the lower board, so that it has soft start using a capacitor, and then regulates the voltage on those capacitors.
@@SeanBZA ah yep, that's does make more sense. I did wonder what that was in there for. In VFD's sometimes a single SCR is used for regenerative/dynamic breaking (3phase motors), but that's a totally different application.
Another use for a three-phase contactor on the front end is for rectifier protection. If phase monitoring determines a phase was lost on the input, the contactor opens to prevent over-current in the rectifier. This is common in large UPS systems.
@@scottwolf9914 thanks for sharing.
Nice!! The piece you were saying was a industrial circuit breaker is actually a big power relay, great Tear Down!!!
Right, make no sense build in an unreachable breaker, when you had fuses they are reachable from the outer side of this case. The correct name for this "big power relay" is contactor.
Nice teardown Dave, that passbank would make a nice showpiece, I was a bit surprised at the low transistor count, but it all adds up. I was curious as to where the igniter was hiding, and it makes sense that it will be in the head itself as it needs to generate around 10kV to start the arc.
Thanks Daryl! Yes I was bit surprised, expected a couple of boards worth, but makes sense when you run the numbers.
A viewer from Czech republic here. I wasn't expecting a shoutout in this badass PSU teardown video.
Greetings from Germany to the Neighbors:)
Schrack is a German top quality industrial relays manufacturer: www.schrack-technik.de/kontakt
I didn't know they produce in your country too.
The world is small:)
@@dieSpinnt Schrack is all over europe www.schrack.cz/fileadmin/f/cz/Bilder/Unternehmen_Kontakt/mapa_2014.jpg
Liquid cooled Argon lasers setup at 90's Raves where amazing and cost mega cash to hire . The laser head would have a seperate box of prisms and mirrors to make the FX . I can't believe this was getting ditched.
Most likely the tube was at the end of its lifetime, and a new semiconductor laser was cheaper then a replacement tube. Which costs tens of thousands of dollars.
most argon krypton or argon lasers are being ditched. and are being sold for pennies or are given away for free. They serve little use because of their requirements and size.. now people want more projectors which even in terms of power and water requirement is just not feasable when going even in numbers of 4 units..
@@Masterpj555 Interestingly enough, those old ion heads had very elaborate optical decks to generate effects which really can't be matched on simple semiconductor lasers with deflection galvos. They basically had to do the same 'wow' factor you get out of a dozen projectors, from a single head and optical setup -- fiber launched split setups did exist but they were not common (and cumbersome to set up).
Pink Floyd was infamous for having very impressive laser displays, which were all controlled with analog multi-track tape players and even synthesizers.
@@Stoney3K
Infamous? Meaning well known for being wicked, disgraceful, evil, despicable?
That's not my memory of Pink Floyd...
@@Stoney3K wow ! had no idea how Floyd controlled the lasers but now it makes sense ! remember the lighting tech/roadie lighting his cigarette off the beam at the start of the delicate sound of thunder concert ?
Awesome Dave! Let's see the inside of the Laser head!
Dave, I was a physics research student and was lucky enough to run my research project using a brand new Coherent Micra oscillator and legend elite amplifier titanium sapphire set.
Just a few weeks before my thesis defense I was running some tuning and managed to accidentally miswire a high voltage supply into the fast photodiode detector port on the controller board. Needless to say, I let the magic smoke out and delayed my graduation by a few months whilst I got it repaired.
I love the channel, keep up the great work!
Hi Dave! In modern labs you won't find Argon Ion lasers anymore. DPSS are commonly used these days, like Coherent Verdi. Better efficiency, single phase powered with the same beam quality.
They're still used in the semiconductor industry, from what I understand. Both Edmund and Coherent still make them.
we have an ion laser exactly like this still in the lab
@@sudazima Microscopy lab? That is certainly an application where it is still sometimes used, although even there they are slowly replaced. There is a reason why so many fluorphores are 488nm
@@leothecrafter4808 no PLD lab, the fluoresce MS has simply laser diodes witht filters
For more demanding outdoor applications like Laservision is doing, DPSS isn't used, instead, Coherent's OPSL is the tool of choice. Even though the beam quality may be *slightly* less than an ion, the logistics of just plugging in a unit on the mains and rigging it like a regular lighting fixture really outweighs the hassle of running all the infrastructure for an ion head.
Ran a few of Coherent and large frame Spectra Physics for entertainment in the mid 90ies. The heaviest part to unload were the transformers to bring european 400V three phase to american 208V for midframe ion lasers or 480V for the large frame SP 171's. We always tapped the transformers so that the voltage across the passbank was just high enough to be in regulation, as not to stress the power supplies.
We were using mostly used ion laser heads from universities that did not meet their specs for the scientific work anymore, but that meant that these heads tried to kill the power supplies by having low voltages across the tube and running at higher currents for acceptable output. So we were used to blown passbank transistors and repaired many exciters on the road, just before a show was about to start. The older generation power supplies without microcontrollers were much better in that regard, with a real knob for current, adjustable magnet current, manual ignitor with unlimited retries to strike the tube and analog meters on the front.
LOVE AvE's Sticker! You have to get it going!!! Dave you must have some 3 phase somewhere!
Nope, no 3 phase
@@EEVblog I see a rotary phase converter in our futures
@@dmacpher Maybe a VFD?
Yeah I really want to see this thing repaired and working again
@@toddberg3892 VFDs don't work with these PSUs for the most part as they expect an inductive load.
Love the AvE sticker
This guy is everywhere 🤣
Is that a real safety sticker :-)
If ever there was an appropriate piece of gear to attach that to, a 10W laser certainly qualifies.
Yeah, he knows what's up. A 5-10 watt laser is enough to destroy you. Will never see any thing again, ever.
Why does everyone think this is an AvE sticker?
The saying originated in Australia long before anywhere else.
You know what the pass transistor array needed? More warnings about high voltage. I don't know about you but when I see huge traces, lots of component isolation, and monster AWG cables sticking out of it, I just want to touch everything. Only having nine warnings is playing with fire.
Well buddy, I high suggest finding yourself a nice pair of electrically insulating gloves. That or learn to poke-prod with a nylon/fiberglass rod.
@@eideticex Go with the gloves! Ya just can't beat hands-on!
YEs, yanke protection must be strong in those kind of products ! XD
Love the "DANGER" sticker on the laser. I have the same one on my table saw
Looks like a sticker from AvE lol
I use a bigger, better one (a "Coherent Innova Sabre") at work quite regulary. It can do multiple wavelenghts and can be used both in multi-line and single-line operation. Tube runs at around 500V at up to 55A, so 27.5kW into the tube, around 40-60kW from the mains. Output in single-line operation depending on the chosen wavelength around 2-12W.
To make it more fun, we use it as a pump light source for a dye-jet laser, which itself has somewhere around 3-20% efficiency.
For your information:
"LIGHT" switches from Constant Current operation to regulation of optical (light) power. At power-up you typically raise the current until you are roughly at the desired power level, then switch over to "LIGHT" and set it where you want it.
"PWR TRK" is "Power Tracking". It's a set of actuators additional to the vernier screws for the back mirror, which does continuous fine tuning for maximum power. This usually is an option you have to pay extra for, but the button is always there.
The problem with those lasers is, they use up their gas during operation. To compensate for this, the tube has a reservoir and a valve to refill when necessary. The problem is, the valve isn't perfectly sealed, so some of the gas is diffusing into the tube all the time. If you leave them powered off for say 2-3 years, the pressure will be too high for the igniter to ionize the tube and you will need a new one.
Ahh, the good ol' days. I can imagine the young blokes in the workshop standing over the power supplies soldering and routing and adjusting the metal work on a grinder, screwing and testing - days of assembly work from beginning to end.
Dave, I hope you tear down the laser too.
Oh SNAP. This is awesome. That relay is sick
Gear!
Hey thought you'd show up here somewhere!
Also would have been nice to even just that which those big old Power resistors you know the aluminum case ones.
I have heard on some of those if you put enough current threw them under right conditions they go off like a shotgun shell.
I've heard people doing that deliberately girl scared the bejeebers out of someone.
Or like those other times where you stick some sort of power resistor or something across the power on a big power supply unbenounced to someone else.
Anything that would let out smoke a nice bag Etc.
Never done it myself but I've heard about people doing it.
Someone told me personally today at dumb it to an instructor setup.
Between classes before his class don't know how he got in there and accomplished it but he did and.
Let's just say dad instructor.
Which I meant later said that was probably the most memorable set up he ever had.
At least when the biggest things that happened to him personally in his teaching career.
You remember when somebody goes off like a shotgun shell that you don't expect to have that happen.
Sort of like when a bag of filter caps all go off at once due to some sort of failure in a power supply.
Only had that happen once but remembered definitely remembered.
Bad shutdown circuit over-voltage by at least double or more!
All it was was one bad resistor took out the shutdown circuit no Fail-Safe
I used for work at Spectra Physics as an engineer working on DPSS picosecond/femtosecond lasers, and in college, I used to work as a repair technician on mostly ion lasers from both Spectra and Coherent. One thing of interest is the concept of power mode and current mode. I'm sure "light mode" puts the tube in closed optical loop operation where an attenuated and properly diffused optical signal is referenced by a photodiode to put the tube in close loop control to maintain a stable light output level. I had a lot of fun applying (discrete-time) control theory to lasers. Using microcontrollers affords a lot of flexibility. At the big convention centers, the sudden inrush of thousands of people on the floor changed the local environment that we could see it in the data-loggers for the control actuation signals, and we were pretty sure that a lot of other vendors were affected as well. We even used water-cooled power meters to verify the power stability. One thing you have to be careful with with closed loop operation is guaranteeing that the beam pointing does not wander, which becomes harder when there are more optics in the box. It was solved for the laser I worked on, and am curious about how this is handled in general.
I love the proper application of AvE warning stickers
Why did all the nightclubs in the 80s and 90s have a swimming-pool ?
It was actually the water cooling for their laser show ! :-)
Yeah and probably definitely enough to keep the swimming pool warm enough to be a therapy pool if the show went long enough
Those gas heads were the cream of the crop for laser displays for years, in particular the Argon-Krypton mixed gas variant which actually produced a white beam, which could be modulated with a PCAOM to select colour. I used to tinker around with a 1980s Spectra-Physics model 168 which was a few watts of white lights out, and about 15kW of electrical power in. All of the excess heat was dumped into the cooling water, not just from the head, but also from the (linear) power supply. The regulation in the PSU was still done by a tube!
I also had an ILT 5940A ion laser head which was air cooled on my hobby bench.
Hello Dave, how are you doing? Just wanted to thank you for all the stuff I've learned from your channel. This is just a tribute :)
I absolutely love the operator warning on the side of the laser. I need to get some of those stickers. Anyway, repair it and turn it on!
I bought a HeNe gas laser probably 25 or 30 years ago because I wanted to get into making holograms, but due to procrastination, lack of money, and lack of a good place to set it all up, I've never actually started it. So, the laser is mounted on my telescope for -nefarious purposes- ...I mean, for staking flags on my property lines between pairs of survey markers.
Czech viewer here! Thanks Dave!
I get giddy when you breakdown stuff from the entertainment industry. when times were better I work in sound and light for entertainment purposes. As times goes I have been learning so much about EE in general. you rock, Dave!!!!!
at the end of this video... that was wonderful! double the thanks.
At 29:48 that's not a circuit breaker, but a contactor (big relay with current break capacity). Software probably starts up and completes power on self checks before pulling in the contactor to apply power to the output stage.
Daryl is good people! He and I spoke and I thoroughly enjoyed the podcast. Sad to see that amazing laser torn down :(
If you ever need a person to interview on laser tech for a Podcast do find me!
The plasma tube is filled with about 350 Millitorr of Argon. The arc has a negative impedance characteristic that makes designing the control loop very interesting. The tube will take all the current you can feed it, and will try to "run away" if the current is not limited. I've rebuilt the tubss and power supplies for those and Coherents competitors for about 30 years now. You'll find the hot cathode in the plasma tube needs around 28 amps at 3.3 volts give or take. The secondary of the cathode transformer is center tapped, and the tube current flows off the center tap onto the high side of the passbank. The transistors almost always fail shorted, so there is a diode and fuse in each base lead and typically a fuse in the collector on the 36 transistors. The 300 series is super sophisticated, an eeprom is created at the factory with the i-v characteristics of the plasma tube, and if the tube doesn't match the eeprom, it warns you for the need for a gas fill or shuts down. The plasma tube you have does not contain Beryllium Oxide, but the insulators under the To3s almost certainly does. I have the field service manual, but I can't post it.
The one thing you're missing is diode lasers are always switching you cannot run them 100% duty cycle (then) 90s. 1950s lasers still existed to make light shows look good on ccd camera. With polarizers for color and mechanical mirrors... And by the way you meant collimated. Luv ya dave!
I had a very small lasershow business in the 2007-2015 period, that was the time were high power laser diodes started to appear. I saw the business of these things basically being destroyed in few years and now you can buy a 10W green laser projector (the whole projector, not only the laser) that is half the size of that power supply for 1/10 of the price of that laser module.
One club where I worked still has one of these beasts. It was still in weekly use when I left in 2015, I would love to go back there and check it out now 🥲
I would pour that block of transistors into resin as a piece of art :)
This takes me back I was a physics postgraduate student and spent two years with a similar unit the Innova 300, think max power out was 30w but we never got more than 24 out of it. My first job was to plumb the water cooling but opening it up by myself was definitely a no no! One of my supervisors had to tune it when it was feeling a bit crook. Total beastie!
Who else had a smile after listening that the PSU is a giant pass transistor?
The fuses on the passbank are there to prevent a fire. There's a huge amount of energy flowing through such a tiny box. I remember swearing at the designers as a technician in college, only to go to lunch with some of them when I joined as an engineer. Ion laser supplies often fail pretty spectacularly, especially the early designs .. Tubes going up to air for example, created a complex impedance that caused a reflection back to the power supply. Please note that the passbank has one set of transistors for the tube, and one for the solenoid magnet to focus the beam down.
I feel like the "disadvantage" of the overall current dropping if one transistor blows could actually an amazing advantage, because the monitoring could basically divide total current by the requested per transistor current to get live feedback of how many transistors are actually working, and therefore how many have blown. "Dear maintenance staff, 2 of my transistors have blown, I can still compensate and operate just fine, but please replace." Of course, this is all speculation.
Also it could tell you WHICH transistors to replace.
@@VintageTechFan How?
@@TheHuesSciTech Well .. I was a little fast here .. you actually need to measure individual current, too .. but since they are all tied to a common potential, a simple analog multiplexer would be enough for this.
The large white block inside, next to the caps looks like a 3ph contactor. Probably bought in after the caps have charged through a resistor. Maybe that weird looking one on the underside you removed?
A lot of large motor inverters use this method also. I have some broken IGBT modules I keep meaning to send in ~1.7kv 450A, you can remove the top and see the silicon components/bond wires through the gel. I think the max thermal dissipation is around 2KW per package.
17:34 Mid-century elegance -- neatly striped Allen-Bradley resistors.
Wow - one of the best 4K videos you've done! Thanks for the detail!
LOOK AT THAT BAD BOY.
97??? Fantastic Vintage😁
Dave??...Your A Highly motivated Peeps generally!
So, meeting Peeps like Darryl, does it give you a spring in your step? And keep things running positive? Cos even if a Peep has everything , they still could feel, like it's same old, same old?? I just find it fascinating how you've moulded your lifestyle to work for you. When most of us feel like we don't hold all the cards.
Well done.
The breaker thing is likely a contactor. Looks like one I would see controlling various HVAC related motors.
Incredible! What a great insight into something you certainly don't see everyday!! Love these unique teardowns Dave.
or maybe VirtualGround?
I'm jelly. Such a beast of a gas laser.
Kanthal is a real popular resistance wire, for making resistive coils in the Vaping industry.
We have a similar one in storage room. Surprised the power supply is so simplistic.
The SCRs are probably for crowbar protection against overvoltage. If the voltage goes over the critical limit, the SCRs turn on (and latch), grounding the outputs, which blows fuses / trips breakers. Might do a little damage in the PS, but not too bad, (something needed to be repaired anyway!) and saves the laser. Was just talking with someone today that told me he lost a radio recently due to his PS shorting a pass transistor, which then dumped 32v into his 12v gear. Wouldn't have killed his radio if the PS had included a crowbar. I know all three of my big PS have crowbars.
I think AvE got his hand on it first.
I still use one like this, but slightly older Zeiss. One thing about this type of laser - there is absolutely no way to shift the lines and they are extremely narrow. When I need to double check my equipment alignment there is nothing to replace our trusty old argon laser. And due to it's purity it's easy to filter it off, so it also works well as a source of excitation for some of my materials.
At 22:17 it looks like one of those transistors may have blown up too. On the row closest to the board edge, four away from the backplane connector.
That was a fun teardown. Astounding inefficiency in that system. Now I need to see if you tore down the head too. It'd be neat to see a it powered by a neon power supply (if it'd even work a little) to see that massive tube glow just a bit.
I remember an old Unholtz-Dickie shaker table amplifier, rated something like 10kW -- poor thing stood three or four racks across, huge 3ph transformer in the bottom of one, bunch of fancy calibrated circuitry in the control unit (for displacement, acceleration, etc. control, sweeping, etc.). The power output I believe used a Circlotron architecture (tube aficionados know about that one!), and consisted of a Metric Shitload of stud type NPN transistors on a common rail, for each output. The rail being a water cooled tube.
Have a few Coherent's in my lab. A 1W CW, 500mW CW OPSL and a 2.5W Q-Switcher, pricey kit but really good for the money. Gonna echo @Filipamator 's comment about how most lasers are DPSS or OPSL technology these days. Also currently working on getting a 20W Verdi. M2 on the CW lasers are 1.02 on the 1W and 1.05 on the 500mW unit, so solid state can definitely hang / surpass these days.
Brainiac would have loved testing it
I think I’ve been waiting all my life for this taredown
I'm loving the "I'm pretty confidant I can do..." as you oh so delicately tap the caps, lol. Be honest, you weren't that confidant (and to be honest I wouldn't either; you never know if a resister has failed)!
I'd be very very confidant. Of course, behind the scenes I'd have clipped a 470k ohm resistor across the capacitor leads half an hour earlier, and only removed that resistor maybe a minute before I started filming.
I am green with envy. Like the green light you might get out of a badass Argon laser.
It looks like those fuses on standoffs would not be fitted initially. The standoffs would be used as the measuring point for the current. The fuse would replace the ammeter once current was set for each transistor.
⚠️don't look into laser beam with remaining eye 👀
* with remaining head, I guess.
Used the larger Version of that with up to 25 Watt of multiline output to pump a coherrent 899-29 single mode dye laser and a Spectra Physics 171 on the same table for a second 899... Had to build an own heat rxchanger for the old stuff, the Coherrent Sabre had its own. The dye lasers were used for spectroscopy of liquid Helium cooled samples at 1.5K. We got a New 100kW power line for that Lab ;-)
The white industrial block thing is not a circuit braker Dave! It is a huge contactor/relay. They brobbably can kill mains power electronically if they need to, or ot is there as a precharge stage together with a resistor to precharge the huge smoothing caps to limit the inrush current.
Dave gets so excited!
33:10. When your hardware turns in to vaporware.
I've dealt with several permanent laser shows around the US. we had a pair of Coherent 80w krypton/argon white light lasers. Their product is second to none!
Our power supplies were some 235 pounds each.
Love it 😎
I have a few of these.
Spectra physics 168 Argon Krypton mix.
A 2020 and a coherent sabre.. it's a beast 30watts.
208v or 480v 3phase input on these so you need a giant 3 phase transformer aswell.. lol fun times
I think I can make ALL the cats in my neighbourhood go crazy with that lazer..
I believe they would had used these for the blue beam, a semiconductor laser for the red beam and a copper vapor laser for the green beam. I didn't know this until recently but they actually make even more powerful argon lasers than this.
Argon has some green lines - if you use an argon/krypton mix and suitable mirrors I think you can get RGB from one laser
Yep, Mike's electric stuff is right. A Krypton/Argon mix will do multiple wavelengths across the spectrum from red to violet, so-called "white-light" ion Lasers.
Dave's new Christmas light show is going to be out of this world!
Maybe the rows and column transistor array is supposed to allow like for 1 or 2 blown transistors and still work. Like if a column lost one they could have been running underrated and just crank up to the same level using the remaining ones.
You should get maintenance to install an outlet in the lab that can support that plug, we need to see this powered up! :D
Hi dave!!! the white dust is simply galvanic corrosion eating the zinc plating .
These things would operate in current mode, as you describe, but also in "light mode", where an optical pick-off in the head would feed back light output to the psu (using a pin photodiode) to give a very stable output, hence a lot of the extra complexity I guess (ie it is constant current AND variable current)
The water cooling cooled the passbank, then went out to the head, came back to cool the rectifiers, then exited via the "drain" , normally to the input of a heat exchanger, where it would be cooled before being sent back to the passbank input. Water bills get high fast if you use the drain as an actual drain ;)
the white stuff is aluminum oxide, most likely from moisture, probably a cooling leak?
In the well-funded lab where I did my PhD, we used fiber lasers with harmonic generators to replace this sort of laser. Much more efficient. Each one cost about $70000 though, and you needed three because they were not really tunable. They did have very good beam quality, as they were in the end solid-state lasers not diode lasers.
That power transistor amp would make a great basis for a awesome audio amplifier ;)
13:10.....Transformers, Robots in Disguise 🤖🤖🤖
19:59....I've heard of a 9 Bar!! But never heard of a 4 Bar?? Hmmmm?? Bueller ?? Bueller??😂
.1 percent efficiency ? Solar roadway : "hold my beer"
This was hilarious :D
That white dust caked on is likely zinc galvanized metal oxidizing over time to zinc oxide. Benign as long as you don't heat the metal up to welding temperatures or try snorting it....
Are you sure the 4 screw devices you identified as bridge rectifiers are actually bridges? I used similarly packaged darlingtons in the past. They had great specs and current handling and they were used in dc motor control devices.
I'd take a guess and say that they could be SCR's in a controlled 3-phase rectifier arrangement, much like the type you would find on variable frequency drives.
@@Stoney3K If that’s the case shouldn’t it have 3 devices instead of 2? That might be a good guess though, those packages puzzle me, I’d love to see what they are! @eevblog any chances to get some pics of the part number?
I'd take the EDLC off the mainboard and save it for another project.
The reason for the high amount of insulation isn't for the operating voltage, which is around 260 volts, but rather the ignition voltage which is around ten kV.
It would be great to see it all working !!
Very cool video Dave. Thanks for sharing.
You overlook The Big Bang on the transistor array :)
i found a Coherent Eye Surgery Medical Laser in the scrapyard a few years back. Best $50 i ever spent. It had a nice blue/green argon laser. Also heavy duty block diodes and mosfets mounted on a liquid cooled block. The red HeNe aiming laser was smashed unfortunately. The interweb told me to be careful with the ceramic laser tube and not to break it or make any dust because its toxic Beryllium Oxide !!!! The lenses and mirrors are possibly sapphire ???? Cha Ching $$$$
Dave I bought a ionization UV machine. I think it was a knock off from China? It has a decent circuit board on it with two high voltage Transformers. It's called "New Age living C 3500 high-efficiency air purifier"
You definitely do not stay in the room with this thing on!
It was covered with what looks like the white salty substance. I found out while cleaning, it water soluble. Much easier to wash off than to scrape off!
Pftt ! I use to be part owner of a laser cutting service. Used a 4KW Trumpf CO2 laser. That's 4KW of actual light output. 10s of KWs going into the machine. Needed 440V 3 phase power, with 250A breakers. Could cut steel 3/4" (18mm) thick !
this takes me back
I had one of those puppies in my Lab (albeit the more powerfull version with 35W all lines and 14W for one line). But i replaced it also with a diod pumped disc laser (with two times 20W laser power (one line) and less than 4kW electrical power usage. Those old Argon lasers are really power dumps ;). the power supply of our coherent argon laser was about 3 times the height and was impossible to lift even a bit (i could only roll it). and there was a second similar sized unit for the cooling (mind you that also needed an external cooling water supplied, just you need it very stable for the laser, that is why it neede the seperate cooling unit. I think power needed was 56kW? but i might remember it wrong :/. We were using it for an experiment in a synchrotron (with a peak power usage of about 3MW, so the laser was just one of the small fries ;) )
That would make a nice monoblock class-A amp :)