EEVblog 1572 - Haefely 16kV ESD Tester TEARDOWN + Repair Part 1
Вставка
- Опубліковано 11 чер 2024
- Teardown and Part 1 repair of a Haefely Trench PESD1600 16kV ESD tester.
If you can help identify the rotary encoder at 25:52 that would be awesome!
Unboxing and testing: www.eevblog.com/2023/10/13/ex...
Forum thread for the rotary encoder: www.eevblog.com/forum/repair/...
Forum: www.eevblog.com/forum/blog/ee...
00:00 - Haefely Trench PESD1600 16kV ESD gun
02:09 - The Onyx16. These things aren't cheap...
02:33 - DIY ESD Gun
03:00 - Teardown
04:08 - Probe tips
05:06 - Inside
05:46 - High voltage relay and pulse shape
07:19 - Front panel PCB and ANNOYING construction
08:09 - NiMH battery pack
09:35 - Did I mention the ANNOYING construction?
13:16 - It's OUT!
14:36 - Main PCB soldering
16:56 - This rotary encoder looks SPECIAL
18:13 - Testing the rotary encoder
20:27 - Are you sure I didn't mention the ANNOYING construction?
21:25 - The special snowflake rotary encoder is OUT
25:52 - HELP! Can you identify this rotary encoder?
If you find my videos useful you may consider supporting the EEVblog on Patreon: / eevblog
Or with crypto:
BTC: 33BsprBQNBtHuVzVwDmqWkpDjYnCouwASM
ETH: 0x68114e40ff4dcdd384750500501e20acf3875f8c
BCH: 35n9KBPw9T7M3NGzpS3t4nUYEf9HbRmkm4
USDC: 0x68114e40ff4dcdd384750500501e20acf3875f8c
LTC: MJfK57ujxy55su4XicVGQc9wcEJf6mAoXF
Web Site: www.eevblog.com
Other channels:
EEVblog2: / eevblog2
EEVdiscover: / eevdiscover
T-Shirts: teespring.com/stores/eevblog
#ElectronicsCreators #Teardown #ESD - Наука та технологія
I found it on german ebay and posted the link the forum, hope it helps.
Godspeed
The rotary encoder looks like it is an EBE BG 16 (Angle pulse encoder), which can be found for example at bürklin electronics
Damn, you seem to be right! 🫡
Wow, that genuniely looks like the thing! Sadly, the data sheet doesn't quite make it clear what the second set of outputs do (at least to a non-EE like me), but physically that pretty much looks like a perfect match.
Edit: From the video, I'd expect one set to "beep" at every stop, the other only every Nth stop. Still, the data sheet doesn't mention anything like that.
@@fonkbadonk5370 The encoder basically connects Ak to common, while between notches and Rk, while on a notch. There is one Ak for each direction.
94 Euro. Holy moly.
@@fonkbadonk5370There are 3 pins for each rotation direction. Per click it connects the output pin to Ak and then back to Rk and like this generates a pulse per click.
I swear I've seen this encoder pin-out in an automotive application before, it looks a lot like the type used in mid 90's OEM stereo head units with that small form factor and "decade" pins for faster steps when tuning the radio...
We have one that we bought new in 2021! It is a different brand, but looks exactly the same as far as I remember. The built-in battery of ours is lithium instead of NiCd, though. And the manual says that if the internal battery goes completely flat, you have to send the whole unit back to the manufacturer, because you won't be able to charge it anymore 'for your safety', of course! Seems like this older model lacks this super handy 'feature'.
*EDIT: UPDATE* 19th NOV, 2023! I checked it about 10 days ago, and it is a SCHLÖDER SESD 216, produced in 2021 I think. It looks 100% identical, even has the same blue ribbon cable ground wire. I don't think it uses the same encoder, though. The knob has a super tiny hex screw to hold it which I did not have a key for, so I couldn't check the shaft.
And guess what: the battery was flat! Fortunately it behaves better than what the manual suggests. I was able to charge the battery. So it has a dual cutout system, just like a phone: below a certain voltage it will refuse to power up, but the cutout voltage of the BMS itself is under that threshold. Probably it will only refuse to take a charge if the BMS is shut down due to undervoltage. So they did it right!
My guess is the unit is very specifically calibrated with the standard battery and a user replaced battery can either void the certification or produce different unwanted pulses that aren't certified.
@@Haarschmuckfachgeschafttadpole There are absolutely zero technical reasons for a non replaceable battery in this case. Its just a method to increase the income of the company.
@@Haarschmuckfachgeschafttadpole I haven't mentioned battery replacement. If it just goes flat because you forget to charge it, it will lock you out, despite the battery is perfectly fine, just needs a charge!
@@mrnmrn1 I thought Apple had a patent on that technology (edit - I think it is limited to telecommunications devices).
There's a little more to it than greed when it comes to Lithium batteries. They have a known failure mode that could happen if deeply discharged.
The fun you're having explains (to me, at least) why you got it for cheap. People see "faulty" and run away screaming.
I can't talk .... I get sucked into the "simple fix" mentality on occasion.
I used to use one of a similar model, a Thermo Keytec MiniZap, back in a former life. I had to use it to test landline (yes, that degree of former life) phones, having to sweep the device while on hook. I remember one Siemens phone I tested, that had this metallic coating of paint around the keypad area, and this coating would actually hold a charge. The ESD gun would not fire on the keypad, but the keypad would discharge to me when touching the paint...quite a discomfort. We had a bunch of interesting gear at the lab. We had quite a set of Keytec ECAT test boxes in 2 racks, which simulate surges and such, similar to what lightning strikes would do on a phone line. The old 500, 1500 and 2500 series sets, the ones with a solenoid operated ringer bells, would ALWAYS chime the bell when with the surge was fired. As for price, yeah, that stuff was not cheap. I was told that each of the modules, which we had 8 of them, would cost more than a nice car.
Though, later in my life at that company, when I moved to the wireless team, we had had CDMA test boxes that were like $180k each, and we had like 6 of them.
Don't butcher the insert, just insert a bolt and lock it with a nut then withdraw or unsrew the whole lot!
Ive never seen an encoder like that before. If you can get it apart without damaging it, im 100% sure the problem is dirty contacts. It'll just need cleaning.
It will need a replacement, because the contact material is worn out. You obviously have zero clue on how encoders internally work.
@@t0scanellisaucer of milk for table 2 plz
@@t0scanelli but we are not talking about other encoders and how they work.
we are talking specifically about this encoder and how it doesnt work.
I can't believe you desoldered all those switches. The tops of em with the panel besel/metal pops off 🤣
I grew up fixing TV's in my dad's TV Shop. We had tons of fun with the 30kv from the flyback or discharging the picture tubes, (yes, TV's used to have big tubes that displayed the shows). I cannot count the number of times my arm discharged one of those things. I had a chassis out on the test bench and had to look around the back when the second anode caught the top of my head and out my chin. That day I learned a quick discharge was not the same as a constant shock through the skull.
Omg. If a YT comment could enstill nightmares. 😮
Another great, entertaining, in my niche video. Bomus points for more vocal agony expressed without shedding a drop of blood. Thank you.
Get it X-rayed Dave to see what is in there. I work in Xray and modern digital Xray is incredible.
If you have to bodge one in I'd solder a cable onto the board and have it fed outside the case. Solder the new encoder on the end of the cable. Then it doesn't matter what shape or size the new encoder is. You could mount it many places on the side of that case.
I think the inner and outer pins on the encoder are for detecting slow versus rapid movement at different distances from the center of the control.
Yes, seems likely.
Full pack done thank you for sharing watched here
I've had difficult switches like that rotary encoder before. One thing I've done that sometimes helps is to use a pin vice and a small drill bit and drill a hole through the case. I would use a syringe with a needle (a local farm supply store had syringes with large needles for animals) filled with white lithium grease and inject the grease inside. Work the controls to spread it around inside.
Pin vise is wizard magic, brings back the dead, Dave should demonstrate this!
Already got contact cleaner down the shaft, didn't do anything.
@EEVblog dip it for few hours, if you have vacum chamber then use it.
Yeah, build the tap welder kit. I was always looking for that :D
Dave. Try to connect encoder to oscilloscope and look what signal pattern it have on remaining pins.
The only similar encoder with 4 mechanical contacts which I was able to find with quick search is EC18AGA used for washing machines. But it is Gray code absolute encoder.
right to repair for ESD guns!!
Best bug zapper money can buy!
Any one else like the sound of frustrated Dave ?
I like the way his voice goes up in pitch as he gets more frustrated , keep waiting for him to go ultrasonic
although he does get close to the sound of an old school CRT
I`ll keep watching and waiting
All jokes aside ,I would say it was dropped , landed backside down and with the rotary encoder sticking about a meter out there poor design
I work at tektronics, we calibrate these guns.. ill ask around if anyone has seen this
rig up a switch that transfers the working pins over to the other two contacts when held. lol Turn to go down, hold button and turn to go up. Should work. haha
Hi i used to do a lot of ESD testing and assessment on company equipment as part of EMC testing and certification and UK / US post office compliance evaluation. We used Schaffner Equipment which was also very expensive but high quality .We used 7 KV for UK/ European requirements and 16KV for US requirements. The two types of tips/probes were for either contact discharge (pointed tip) and air discharge (rounded tip). Also the simulated human body impedance varied with authority's complliance conditions.
I would recommend 'Dremelling' out the perimeter seal first, and lifting off the 'cover'.
There is a chance that the 4 heat welded pins hold the contact armature 'cage' (as an assembly of loose parts).
@therealjammit had a good suggestion of drilling a small hole to get penetration of a cleaner into the encoder.
I have also used vacuum or heat to suck solvent into a part: I oiled a clockworks type timer motor by alternately placing in a warmed oven (burner off) and cooling in fridge, while the shaft 'bearing' area is immersed.
I found a small spray can cap, or ink marker cap, to hold the fluid, with the part 'upside-down' in the cap.
[shaft facing 'up' during heating, and facing 'down', immersed in liquid during cooling.]
I got so much oil in that I could hear it when I shook it, after just a few heat cycles, (much more than could be infused by oil applied to the shaft). [drilling a hole would have been my next alternative].
You can use the thinnest 'Dremel' saw blade or cutoff disk, or an ultrafine burr you can find; or grind the sides of a stock saw blade thinner. [run the blade at medium speed against a running bench grinder. -- SAFETY GLASSES!]
Pretty much identical to the ones we use in my lab. I always wondered what was inside, aside from the interchangeable caps and resistors.
I double checked to be sure. Yeah, our older ESD guns are Haefly, but they go up to 30 kV instead of 16. I imagine they're stupidly expensive. 😅
what are u testing with these? dont tell me computer motherboards😆
@@echelonrank3927 Avionics. Military, commercial, and space. Also military C3I stuff that doesn't fly. Needless to say, the standards are *very high* for these applications.
I do occasionally do ESD testing when the lab is slow, but mostly that's delegated to the "341" garden-variety techs. They mainly reserve guys like me ("170") for when tests must be conducted that are hazardous to the equipment, where there are clear standards but no set procedure. They trust us to get accurate and repeatable results without breaking stuff in the process.
It's a pretty cool job.
@@GoSlash27 i have a few european tram motor drive systems at work built to withstand lightning hits. but not australian heat 🥵
@@echelonrank3927 Yeah, lightning testing is a whole other beast. We use Thermo Scientific ECAT units for that. I've never checked to see what they cost, but it must be insane.
Hi Dave, when I was studying electronics engineering at technical college in the 1990s, the CE marking requirements were introduced so it became part of the curriculum. We students were given the task of developing and constructing all the necessary test equipment.
My task was to develop an instrument like the ESD gun you have just examined and because Greenland is part of Denmark I also had to make a 25kV version, because it is demanding in arctic countries due to the very dry air that personal discharges reach that high voltage.
The anonymous IC may be a SAB80C535 they were popular back then.
Dremel the sides off and stakes, all ways 3d print a new one.
I think its custom made for that item, no markings.
Hey Dave you should put out a multimeter with a video output. Then UA-camrs wouldn't have to mess around pointing a camera at their meter.
Had an opportunity to disassemble a Schafner esd gun, and they are not easy to disassemble at all.
May want to look around for a service manual, possibly the manufacturer has a full parts list to help the part search.
Drill out the black dots on the encoder. Those likely connect to the “outer” front case, and slides through the encoder to the back, and then are melted to lock the case. If u drill those black dots, the encoder should come apart.
Hopefully you can fix a replacement or fix that encoder. I would hate to have to see a second one of these guns get taken apart for the encoder. What a PITA getting the thing apart!
Just a thought. A desperate solution. Use a toggle switch and switch the output of the correct side (up) to either up or down input of the IC.
5:40 if we're in, but not in like Flynn, that completely destroys our confidence!
That Jamico cap looks a little bloated....I couldn't tell you how many bad Jamico electrolytics I've replaced over the years. If you manage to rehabilitate or replace the encoder, you might also consider checking caps while you have it apart and replacing any that are suspect.
dont it say jamicon? i found them surprisingly unshit in things.
@@echelonrank3927 D'oh! 🤦♂ Still, from my stint with repairing consumer audio gear from the early '00s I can tell you that in my experience they seem to have a high rate of failure.
I've use their test equipment for years and have always pronounced it Hayflee.
Cheaper to build one your self. Long time ago I needed to test 10kV caps. I took the flyback of an old TV, coupled it to an oscillator and it made up to 40 kV.. Would be more work to make the correct pulse shape etc but it is do-able.
At 17:00 on the PCB it says (in german) "Solder Side ESD 2002". So I guess it either already has a replacement or upgrade board or is built up with very old stocked parts (or whole assemblies?).
How about you bodge wire the down pin to the up via so that you can get it up and with reset ist already starts low. So that you can do some sappin right away?
I'd say to bridge the functional pin on the one side of the encoder to the non functional pin and hope that increments the value without too much issue.
Was about to try that, but I soldered it back in and the good side is now completely faulty on both output pins!
@@EEVblog :,( sorry for your loss. though would love to see another video of the process of hacking together a replacement if no matching part shows up
@@EEVblog Dave, check out the Schlöder SESD 216. It is still in production*, and looks 100% identical.
*: At least it was still available in 2021, when my emplyer bought it.
I have a feeling that unmarked PLCC MCU is a MC68HC11.
It smells like a Motorola.
My first thought was MC68HC11F1. I remember writing custom bootloaders for them.
Could be a rotary switch?, I see some encoders made with 4 positions rotary switchs
Just dip whole encoder in contact spray for few hours.
@@EEvblogofficial001.Fu, dave please block this scam person
To open it, you need to use bit made for milling machine that has flat surface tip. You see four points where ultrasonic welding was applied. Drill it just to the depth of 1mm. After that it is ready to be opened.
I have come across them but never used them in circuits, but i thought the encoders need a voltage to work?
Five pin encoder with an extra earth. Is that a regular encoder mounted in a magic box to fool the muggles Dave? ⚡
Great work, I'd like to see you take it apart to see what's going on inside. You could do it with a dremel with a narrow saw blade or grinder. If you mail it to me I'll do it, but I'm in California.
What's your TW?
I like looking forward to 2nd episodes.
lightning strokes on the website.. yeah , i'd have that too..
those look like li-ion cells.
i want one :D
not sure, but the big pins of the encoder look a bit like ones i've seen on some switches. can't really remeber. those could be screwed out for maintenance.
If you feed the pulse it generates up its own cloaca, does it pass the test?
Is that a nice ESD tip or is it just happy to see me?
Stop You're making it blush lol
I can feel my pants getting tight already.
@@markm0000 Maybe youire just eating too many donuts.
@@ztechrepairs True
@@markm0000 😂😂
Find a suitable encoder, without the same pin placement. ( same function )
If you can find one that fits the placement and the pins don’t match the same pin holes.
As long as it physically fits the location you can wire it to the holes.
( using small gauge wires, from pins to board holes ) ( cut the pins short )
Maybe you have to enlarge the mounting hole, but it will work.
Good luck
I wonder if the center hole on the shaft is open towards the inside of the encoder. If yes, it's spritzen time!
some schnaps
looks like that is a custom jobby you could get e standard 6 pin and use wires to connect it to the board or open it up and clean the contacts seems your only option
Yeah I've seen one of those encoders before. I'm pretty sure it was in a...
Hafely Trench PESD 1600
Sorry, could not resist for some reason :P
I'd try to flood it Deoxit down the shaft....
In my experience of ESD testing - being able to do this yourself is one of the most useful things you can do - being confident your product is going to pass before you get to the EMC lab can save hundreds or even thousands of pounds/dollars, etc.
u probably never had anything fail ur own or any other ESD product test before and yet somehow its still money well spent on the machine
@@echelonrank3927 Before I bought my kit (second hand) I had experienced several failures, notably screens attached to processor boards and unwanted resets. Having an ESD simulator lets you generate events that might cause resets or crashes. Depending on standard and test level, so long as the system recovers properly, it is often acceptable to crash. In one product, I ended up periodically cycling the power on a screen (every 30 minutes) to ensure it didn't lock up permanently - it was a monitoring product used in harsh environments. Test lab costs are typically £900/day in UK.
Half expecting you to grow a uni-brow and start saying FULL BRIDGE RECTIFIAH with an ESD gun like that 🤣🤣
im just a novice at this stuff i would get something with the same output format and amount of pins then just bodge it in. but thats the only way i can see. i mean even if you could find the part number of it, would you really be able to buy it? Id put my money on it being obsolete or proprietary or both.
It has a date code of 2002 on the board at 14:44
Would love to the the real fault inside the switch. ❤
That is one severe-looking instrument.
WHOW a real pest1600 !
this bad boy tests gaming rigs faster than u can say dead dingoes dongle
given the time period, the blank chip is likely a PIC16 something
I was thinking maybe a 68HC11F1
Have you found one Dave? I think ive got one on a old board here i can test and send you if its the same ...
Steve
Take a look a Bourns, redhill or greyhill brands, they produce a dual sweep quad precision encoders which that might be . sometimes they paired with a lm chip .
Doesn't seem to be Grayhill or Bourns
Set phasers on stun Spock! Dammit Jim I’m a doctor not a gunslinger!
1000 th like 🎉 👍👍
Hi Dave,
An incremental encoder with external pullups, strictly, requires ONLY three pins.
This device has SIX pins. (five, if you count the two "grounds" as one.)
It is likely that the device *isn't* broken.
Can you check what the functions on the pins are?
Any other clues? For example, is there a hidden magnet inside the device, and two (of four) reed-switches as pick-offs?
look at the datasheet for the ebe bg 16 and reconsider your comment :)
@@_DSch Yeah - that was still when the part was yet to be found.
Bad design is bad. With six pins, at least you'd expect some redundancy.
Wow did you get that from the Ghostbusters?
I like to think I'm a pretty smart guy, but electronics are just something that my brain refuses to understand.. it's too abstract for me, I think. I'm so glad people like Dave exist, and understand all this to such a high degree. Means I can just click, sit back and enjoy! Even if I have absolutely no idea what he's talking about. xD
Misread the title as "16kV taser"
I originally had G-U-N in the title, but figured after 3D Print General just had his channel deleted because of that, Tester it is.
@@EEVblog Good on ya. UA-cam's rough these days
@@nsf001-3 This website is getting worse every passing year. Yet here we are voluntarily choosing to suffer as the water boils faster. There's plenty of other web video platforms, but none quite like youtube.
Had it off for many hours 😂😂😂
I would LOVE to know why there so bloody exspencive?
Make one Dave, as it'll be a lot cheaper and a lot easier to work on if something were to shit the bed, like this one has.
Haefely AG is really close to where i live, doubt they gonna just give me a spare encoder if i knock on their door doe...
also pronanounciation or how i would say it atleast, easiest way is if you put Häfeli in google translate and put it on finnish, the way it says it if you press the speaker is pretty much spot on
I remember there were encoders from greyhill they where 6pin jobby ones, but they where optical and need separate power for the led. Maybe you should reverse engineer the shematic around the encoder to firstly analyse how it is connected. Otherwise do they have a push-button functionality or maybe do they have A and B connection and also the inverse of it !A and !B?
It's two ground and 4 resistor pullup outputs.
greyhill made good gear. but dont forget, this the super expensive german stuff were dealing with here,
so theres practically no way that cheap crappy encoder can possibly be optical.
Personally i would just grab a regular 5 pin encoder (assuming its a push button type, 3 pin if not) and wire it around, i hoping something works
of course being careful not to short anything out, maybe put a 1k resistor on each pin or something to be safe
Since only one pin pair is active in each direction, the original encoder must have some sort of directional clutch to move two separate 2bits encoders. You'll need a micro-controller or conversion logic to translate a standard encoder to that.
write the swiss and ask for a spare...it will look good on their CSR bottomline
Optical encoder ?
15:18 dry as a what!? XD
Do you have to verbally warn the IC before deploying the Tazer? Just need a voice sample greeting card that says, "Don't Taz me Bro".
If its not optical soak it in contact cleaner. Just dip it. It might be optical. Black pins for led and white for optotransistor A and B.
no way this plastic mass produced miniature junk is optical and besides the black pins both go to ground on PCB
This was soldered in a soldering wave machine those clean solder joints with little solder and the pins getting out with almost no soldering are typical. This is why this was nasty to reflow.
Contact cleaner?
Why not trace the encoderbnets to the ic to see how many channels, possble rc filtering etc. The micro may be a 68HC11. Check the xtal pins
Theer are 4 wires, two for each direction it seems.
Try getting contact cleaner down the shaft._
This is half a Lee ... It does not have the punch of a full Bruce but it still delivers.
Given the shaft, could that be an absolute position encoder that output grey codes?
Its worth trying to get a bit of switch cleaner into the top of the shaft before carving it open. That is less destructive and doesn't cost anything.
Flood it with contact cleaner?
@@EEvblogofficial001. oh wow I've won something, do tell me more.... Hope its not a scam again.
awww no zappy ;(
Ha! So annoying to discombobulate. Desoldering gun to the rescue! Lovely build though.
you need to make you're own and make them in OZ and make the affordable do a kick starter and will all pre order
HUGE market!
@@EEVblogYou'll need to design and build a test and calibration fixture, which may be way more expensive than the gun.
Just soak it in deoxit and turn it. Worse case drill the melted points and take apart to clean. Most likely just dirty contact points.
its epoxy sealed around the perimeter
It should be able to soak in past the shaft
PCB said 2002 didnt it?
should have used your big knife. missed chance! :)
Never did like the usage of, "If you have to ask you cannot afford it..." - it's flawed reasoning. So....what's worth 5-figures on this thing?
Don't zap yourself with this puppy fiddling around inside laddy