just wanted to say I think its pretty dang cool that you take the time to make these videos and share your knowledge and personal perspective with those of us floundering through the forest in search of lupins..
Yet another educational and entertaining piece of work that you did Tony. Many teachers could learn from you how to combine these two elements when explaining a subject, whatever that subject is, not only electronics. What I personally find very useful in your videos is a logical structure behind what you present. Starting with explanation of the idea and theoretical background, then diving into a practical application and finally demonstrating certain number of tests that not only prove what was said before, but also helps to better understand that theory. We know you don't use script. You simply don't need it. Excellent job.
A much more elegant and easy solution for inrush, is to use a NTC thermistor device on the line. We utilized this method on all of our high current amps at Altec-Lansing, which included the EV line as well. - K5DBX
Excellent! I just subscribed after seeing this elegant solution. I build tube amps and I typically heat the early stage filaments with DC to limit AC hum. To do this you need a lot of capacitance because 4 tubes can draw a couple amps and you need that capacitance to keep up with the ripple. But this does present a big inrush strain on the filament winding of the transformer. For 6,3 volt filaments maybe I'll experiment with a 5V or 6V relay. If I can find a sweet spot, I'll definitely add this to all my projects. Many people use a 555 timer or such, but this is downright simple with far fewer points of failure. Thanks!
Thank you for the added information displayed in yellow text. That is always helpful - especially when dealing with numbers, values, and equations. I usually take a screenshot of that so that I can study it at the end of the video. Details matter. 👍
Hi Tony. If you have a relay with an extra set of contacts (DPDT), you can wire in an extra resistor (whose value will have to be determined by experimentation) in series with the one already there, and put this resistor in parallel with the normally-closed contacts so the resistor is shorted on initial power-up. Once the relay pulls in, the N.C. contacts open, and the 2nd resistor is now in the circuit so the relay only needs the hold-in voltage/current. This should extend the life of the relay coil. You may need a small capacitor to prevent relay chatter, though, across the N.C. contacts/resistor. Maybe .01-.1 uF disk capacitor.
24:54 If you want, you can add a small neon bulb in parallel with the relay coil. During normal operation, the breakdown voltage of the neon bulb is way higher then the voltage of the protection circuit itself and the neon bulb acts like a open circuit. When back EMF is generated by the relay coil, the voltage will be clamped down once it reach the breakdown voltage of the neon bulb.
I made a similar in-rush limiter on my DIY variable DC power supply. The input voltage of the supply is 50V DC, rectified from a 36VAC toroidal transformer. I put a 330ohm 5W resistor in series with the 4x4700uF parallel caps, so when powered on, they charge slowly. The relay coil (48V) is connected directly to the caps, the voltage increases slowly as the caps charge up thru the 330 ohms, and when they reach around 38-40V, the relay pulls in and cancels the resistor. Works well, takes around 3 seconds from a cold start. 48V relay coil has no objections to 50V so I did not use a dropper resistor there. Now I see this doesn't help with the in-rush of the transformer, so I need to re-think the whole thing. Thanks for the video.
The only difference I see is that you put the current limiting resistor (330 ohm) on the output of the transformer instead of the input. I'm guessing that you've accomplished the same thing, in that you are limiting the inrush current (to the caps) that the transformer secondary is supplying at a slower rate. Seems good to me. 🤷♂
That a great circuit. I do a lot of power supply's and Tube RF amps that have large capacitors . Them values of inrush resistors can change with size of transformer and voltage of line. with that amp in video you could use 48V relay coil. 24 DC is very common. Also that inrush current is hard on diodes . I had so many go bad . even broadcast transmitters use soft start . just 3 resistors for 3 phase with a contactor across. and the output tubes filament supply has step start . taps on primary of filament transformer. There is a outher soft start circuit that is similar and works great I use on tube RF amps. ham radio type. using a 240V or 120V relay coil across transformer primary. then the series resistor is 20 to 30 ohm 240V 10 ohm 120V 50 watt or so with a 2 or 3 amp slow blow fuse in series with resistor. If you have a diode short or tube short the 2 or 3 amp slow blow fuse blows saving the big surge. I copied the soft start from 3 phase and made it for smaller units. mind you running 480V with 277V to natural. 277V contactor. 500 watt resistors in a cage . 3 cartage fuses 600V . does same thing. large or small that soft start really works. Thank you,😀
Thanks a lot. I am planning to build a power amp using a 625W, 30v 20A (x2) toroidal transformer and 2 large capacitors (22000uF). I was considering using an Inrush limiting NTC thermistor but I realized that during quick on/off cycles may not work well. I also realized that some high end vintage amplifiers (like Nakamichi 620) are using relays as you are suggesting. Unless somebody has a good reason to use Inrush limiting NTC thermistor, I will go with your suggestion. Thanks again.
I like this solution better than the thermal ones. Quick cycling can defeat those. Whether it's someone playing with the power switch or just quick power-outs, it's bound to happen.
The thing that bothers me is that the rail voltage goes pretty low before the relay opens. I'd rather have it open when the capacitors drop to 80% charged.
You could also use a Infineon IRF3205PbF (RDS(on) = 8.0mΩ) power mosfet across the 4R7, rated for 110A @ 55V on the secondary side of the transformer with a 10K / 10R voltage divider on the capacitor. No power resistors needed and no useless heat generated, simulator shows it works fine, it gradually start opening the mosfet.
You make very good instructional videos as your explanations are very thorough. Regarding the Marantz design, I am surprised that 60Volt caps are being used with approximately 50 volts applied across them. 50/60 would be an 83% stress level on those caps, and if it's 57 volts, as you stated later in the video, that would represent a 95% stress level. Could be why one of them failed open. Mil Spec defines high reliability, at stress levels of 50%. I'd imagine the form factor of the caps, available at the time, was the major consideration, but newer caps have smaller form factors and if it were my amp, I'd try to get 100 Volt caps. Also not to nit pick, but you didn't mention that the inrush current also stresses the caps, you mentioned that it could hurt the power switch and x-fmr, (you probably just forgot to mention). Just an observation, not meaning to be critical for the sake of being critical, just providing some feedback. All in all I enjoy your videos, and find them very practical. Thanks for taking the time for the tutorials, as they are probably very time consuming and difficult to pull off.
That big inrush can damage all in its path. A 6 volt relay has a 30 ohm coil if your initial charge is through the relays coil when the difference in voltage is less than 2,5 volts the relay drops out and the Cap is directly connected to the source through NC contacts
Current limiting is essential .the inrush current will damage both the transformer and the capacitors .And since audio quality can be greatly affected by power supply it makes sense to protect your power supply and your equipment connected to it
I thought you would like to know that I’m going to put this soft start circuit you show into an amplifier I am recapping for a guy. I don’t support this necessarily but he has a Hafler DH-120 power amp and he bought o capacitor upgrade kit for it which among other things increases the main filter caps from 6,800uF to 22,000uF. It seems to be a common thing for owners of this amp to do. I advised him not to do it but recommended that if he really wanted it done, he should at the very least add the soft start circuit and upgrade the rectifier. I still couldn’t take responsibility for what this modification might do. I don’t recommend it but it’s his call. Without looking at a schematic, other than start up, would you have any other potential concerns with caps this large put into a circuit that wasn’t designed for it? Anyway, I just wanted to let you know that the videos you put out are a great educational tool for people like me. You are a priceless asset for the community. Thank You
Great circuit addition, however I like to be more conservative and put a capacitor across the relay coil to increase the delay time, and increase the power resistor to limit the in rush current more. I have found that this is less stressful on the rectifiers and main filter capacitors and increases their life.
The purpose of the 2k resistor is to set the level at which the relay comes in. As it is, the relay won't come in until the capacitor bank is charged above 75%. Adjusting the value of the resistor adjusts the trip level of the relay. Adding another capacitor across the relay is unnecessary.
I'd probably just wire a NTC thermistor in line with the mains winding, if I could find anything durable enough for the job. When turning the amp on, this thing's resistance is a few ohms, decreasing as it heats up under current. It will eventually reach a steady state temperature and resistance, dissipating some heat. BTW, the dim bulb is a PTC thermistor.
I do the same thing. I add one in every receiver I restore. I'd be interested if anyone had reasons why using an NTC thermistor as an inrush current limiter is a bad idea.
@@davewhittaker9547 I'm not sure if anyone would argue it's a BAD idea. The repeat power-on problem is probably the most valid reason to not use them. If you've got balance/bleeder resistors across the caps- or any parallel resistance to 0V for that matter- the caps will likely drain way before the NTC cools down to its resistive state after powering down. So powering on the device again would theoretically expose it to inrush current conditions despite having that NTC in there. I know some people might not like them because of their non-linear resistive qualities at operating temp. They'll often bypass them with a relay to take 'em out of circuit once they've done their job. That's kind of a counterintuitive use of the NTC technology in my opinion. If you're going to involve relays and bypass the damn thing then why not use a resistor like Tony's doing here. NTC's are fine, I use them sometimes if space is an issue. But if you're anticipating a problem and plan on doing something about it then you might as well do the best you can within your means. Woa I said a lot of words. 😅
@@deviantmultimedia9497 I see what you're saying. I never thought about the recovery time, but it could be an issue. Not often, but every once in a while I might quickly power cycle it. I need to give this some thought; especially since I also routinely increase the size of the filter/power capacitors near the transformer. I'm curious now, and I think I'll do some experiments on how long the NTC thermistors I use take to recover. Thanks for taking the time to post the explanation.
A straight forward & simple circuit. I.M.H.O. wiring with relay across the entire dc (+/- ....& not +/ ground ) with appropriate resistance and suitably rated zener....this ensures we see sufficient voltage buildup on " both " caps.. instead of only one.. ( eg for +/-24 dc , a 22/27V zener) .. in series with the resistance and relay coil...the coil must have its f.w.d. across it... Relays can pick up @70% rated volts and will drop out when voltage falls below 30% rated volts.....A24 v relay,once energised... will reliabily remain energized between approx. 17 & 9 volts......
I got to say my halfler DH 200 is a heavy hitter my lights dim turn it on so I just leave it plugged in and leave it on all times it is a nice and warm at all times
I'm wondering if an inrush current limiter (like Amphenol CL series) would be an adequate alternative solution. I know that it's a tiny space heater, which is not ideal, but with so much empty space inside it shouldn't be a problem. Oh, and thank you so much for sharing your wisdom, yet again! Please know, it's appreciated beyond words.
I was thinking about a fly back diode may increase the time of the kick out when turning the power off. Isn’t the caps taking the charge built from the magnetic field of the coil? Curious if the relay is actually causing the capacitors to bleed out it’s charge. Am I asking right?
Simple & cheap. Will put its finite hot state resistance there it's the only drawback. Can modulate mains draw a bit along with bass transients if the amp is really pushed. Would be a non issue to a pure class a amp with same power draw at any output level but such beasts are rare.
Nice. Thanks to you i just discovered that my amp has something similar already installed ( Onkyo M 509 ). But it just uses 5*0.22r5w and no resistor before the coil but a cap and a diode across the coil. Now im thinking about changing the 5*0.22r ones against one 4r50w because it has new and bigger filtercaps now too ;) (from 4*30000 to 33000) Oh and its one relais for both transformers.
Could you replace the 4 ohm resistor with an NTC and then instead of using just a resistor inline with the relay coil, add a zener diode. The zener will give better control of the relay switching point and increase the drop-out voltage.
How many amps is the secondary windings producing? And what is the DC voltage after your transformer is rectified? +58-0-58 +60-0-60 DCV ? Thanks for sharing the video great insight much appreciated and I wish you a 🦃 Happy Thanksgiving💯🤗
I've done similar using a scr that slowly (over 10 cucles shorts out a resistor that is in series with the input of the transistor. The only thing agaisnt this is even wen it's full on it will still rob a volt of the incoming votge because the acr does not go to zero.
That is a good idea, but the EMF voltage from the relay coil when the coil is abruptly turned off can be easily over 100 Volts and this would instantly destroy the Zener diode(s) installed between the 33000 uF cap and the relay coil. If you do install a flyback diode in parallel to the relay coil, that will mitigate that possibility. I think that the circuit presented in this video had a purpose of being simple, effective and durable.
@@kylesmithiii6150 Suitably rated Zener diode in series with the 2k/5W resistor...will give an additional hold-off voltage & time delay.. before relay kicks in ..a F.W.D. across coil is a must.
@@NiHaoMike64 Since the voltage on an inductor is V = L di/dt , so if the current across the coil changes rapidly then there will be a high voltage on it (actually a reverse voltage in this case since di/dt is negative). In this case, because the relay coil is in series with a large RC circuit, and C is very large (33000 uF), then di/dt should not be big at all. It can be seen on an oscilloscope. At 27m 57s Tony showed the peak voltage on the 33000 uF, it is about +57 VDC. Let’s suppose that the capacitor discharges rapidly when the amp is turned off and the EMF voltage of the relay coil is 10 V peak, so the 2k (5W) resistor will take the +57 - (-10) = 67 V peak and dissipate the energy rapidly on itself. All will be fine. If you put a Zener diode in series with the 2k resistor (and I do like the idea!), say a 12 V Zener to set the threshold, then first you will have to recalculate the value of the resistor = (57-12-24) / 30 mA = 700 Ohms (2W). Zener value = 12 V (1W) or use 2 x 6 V Zeners in series (0.5W each). You will have to see on the oscilloscope the EMF of the relay coil to see if the Zener(s) can take it, or just install a flyback diode in parallel to the coil and it should be fine. (I am adopting the nominal current of the relay to be 24 V / 840 Ohms = 30 mA) (Tony measured the resistance of the coil to be 840 Ohms - see clip at 15m 56s)
Hello! You always have such great videos! I am learning a lot! Concerning the control of the capacitors and their charging, why not just add a 30 ohm, 100watt resistor on the positive side on each capacitor to limit the inrush of current when powered on?
A good, practical build for a soft start, thanks for sharing it! One thing I wonder about relays that are "soft started", such as ones used for your circuit here or for DC protection on amp outputs, is, does the "slow" build up of DC voltage on the coil to close the contacts make a difference over time to any "self cleaning" properties that a hard "0 to working voltage" application may cause. I'm wondering that, if contacts close a bit harder, will they "self clean" any better.
great question. My gut feeling isit won't make any significant difference in the contact cleaning as the pent up tension in the spring provides most of the cleaning power.
How long is time delay before the relay "kicks in" without the dim-bulb ? From what I've read , you need at least a 250ms relay turn on delay. In addition the larger the trans output voltage and caps on the supply the greater the "turn on delay" should be. I have about 1 sec delay with an effective bleed resistance of about 50 ohms (4 * 200 ohm at 5 watts ea), but the soft start circuit is more complex (complete with a relay turn-on LED) in its own small case.
Can I use soft start from (or designed for) power tools? They seem to be cheap and made in small nice packages easy to fit to audio amplifier. Not sure if I should be trying such devices for audio amplifiers.
Tony, I am curious about your choice of using the 2k ohm resistor to drop the 57 volts to 24 volts for the relay. If one plugs those in a voltage divider calculator with 850 ohms for the relay the voltage across the relay is 17 volts. I’m trying to figure out what dropping resistor I will need to use with the very same relay with a B+ of 70 volts. Using a voltage divider circuit calculator, it comes up with a 1667 ohm resistor. I thoroughly enjoy watching your videos Larry
A little off topic, but how about using a soft start with a SCR voltage regulator, a 100A rated bridge rectifier and 10,000uF filtering caps driving a DC motor?
Wonder if it's a good idea to soft-start a decently sized SMPS like this. Proper ones will have an NTC on the input, but if that's already hot after it's been in use for a while, it can very well act like a piece of wire, passing full current to the main cap(s) and causing arcing across switches/relays. I WAS thinking of using this method, but I heard it's not recommended to "starve" a SMPS of current for too long by placing a resistor/light bulb in series with its mains input, so I guess the resistor value and "hold" time would have to be pretty precise so as not to reduce its lifespan. If it's got a PFC boost circuit it's probably even worse...
excellent work, quite educational. does the loud 'thump' some amps produce when plugged in or coming out of sleep is also the effect of a large inrush current? my bass guitar amp makes one when turned ON and also OFF. as i recall the power caps are way smaller though, i think 2 caps @5600uF. i could try powering it thru a light bulb like shown here.. :}
The thump is because the amp does not have a speaker protect relay. Until the transistors come up to their proper bias level, they could have uneven voltages on the outputs, which causes a temporary dc offset.
To keep it simple how about just adding a CL-30 or 40 or other properly sized thermistor after the fuse ? They are ideal for limiting inrush current and with today's higher line voltages they might drop that a volt or 2 after warm up.
I’m afraid this won’t work in Europe or anywhere else where the grid AC is 230V. The inrush resistor’s power capacity will have to be prohibitively high; we’re talking quadruple-digit wattages here.
Tony, Great information. I repair welding equipment for my living and all of the welding inverter power supplies I repair employ a similar "pre-charge" circuit for the buss caps. I deal with caps at least as big the two here. How do you measure them? I cannot find a meter over 10k. Thank you so much!
Another nice video and explanation but since I don't have a test setup, how much does that 4 ohm resistor reduce the inrush current? And more out of curiosity, how much is 4 ohms (percentage wise) in relation to the power x-fmr primary resistance?
Wanted or not, I will throw in some feedback. (and let the arguments begin 🤣🤣🤣) Believe it or not, you posed kind of a "trick question" , because, since the line voltage is "alternating current" it depends on the exact timing on when the power switch makes contact in relation to the AC cycle. If you switch it on during the "zero crossing" part of the cycle, it would have to reduce less power than if you switched it on during the peak. That, and adding any residual magnetism stored in the transformers body, and also, when the switch energizes it, things get crazy. If you convert things to DC though, you can do some "fixed value" math, and lets say ( ballpark-worst case scenario, an amp that keeps tripping a circuit breaker when you power it on) you have a 15A surge, which, may trip a standard US home branch circuit breaker, at 120V, you would be drawing 1800 watts for a millisecond or so. Ohms law works that out to an 8 ohm load. Adding 4 ohms to that makes it 12 ohms, dropping that current to 10 amps from 15, which is a bit easier on the switch contact, but keep in mind that is worst ( guestimated) case scenario with 60 cycle AC. And, also, keep in mind, that the transformers "resistance", once connected to AC line power, is actually mostly inductive impedance, so, although it may have a lower DC resistance, once it is connected to the AC frequency it was designed for, it will look like it is a higher resistive value than its DC measurement is. That impedance value depends totally on the differing transformers designs, so it will have no set answer as well.
PTC thermistors increase resistance with increase in temperature. You would want to use an NTC. The issue with them is that they get pretty hot when they are at their low resistance level. This adds a lot of heat to the inside of the amp. That's why i don't like to use them.
@@xraytonyb ... and since they run hot, they present resistance, which increases output impedance of the power supply - not a welcome addiition in an amplifier, unless you're looking to limit its instantaneous current delivery capability.
Might your soft start circuit be even more simplified, by using a Mains VAC Supplied STDP Relay? The resistor is wired across the Normally-Open contacts of the Relay. The Line side feed goes through the resistor, to limit the in-rush to the relay and the load. Once the relay and load reach 80% of the mains voltage (80% being the typical voltage at which a relay engages,) the relay engages, thereby bypassing the resistor, and the load finishes charging (but much less violently.) Not my idea = Credit goes to james_s, on the EEVBlog Forum. Thanks, james_s for (originally) suggesting this to me.
Hi Tony…..I’m wondering why you put the 4 ohm resistor on the mains instead of on the output of the transformer. Wouldn’t it operate the same way and still protect the transformer? Just curious. I’m adding one now and would like to know. Thank You
Tony, What is the voltage drop across that 4.7 ohms resistor?. Is that enough to have that same voltage drop in parallel with the relay coil, so that the relay turns off after the capacitors are charged? This way the relay will not need to stay on all the time when the amplifier is on. Would this be a suitable alternative way to do this?
Hi, Tony. A related question, please... I am building my DIY AC PSU chain. It consists of: 1) Emergency All Stop Button to 2) Isolation Transformer to 3) Variable Autotransformer to 4) Dim Bulb Current Limiter (Multi-Bulb) to 5) DUT. Each piece is independent, so that it can be used separately, if need be, for other purposes. Might it be helpful to include a NTC Current Limiter? Advantages/Disadvantages? If so, which component should I add it to? The input on the Dim Bulb Current Limiter? It sort of seems redundant, unless all the bulbs are bypassed... But, I won't always be using the Dim Bulb, in the chain. Thank you for your help.I posted Disclosure> I posted this question on a forum, but I never got a response.
Got roughly the same idea for a mains testing box in my lab, only integrated and not separated. NTC is useful before transformers, especially toroidal. These things can have an inrush current on their own.
@@KeritechElectronics Thanks, Keri. The Isolation Transformer is the problem, it keeps popping the low amp fuse that is in the Emergency Stop Button. It makes sense to add the NTC to the Iso Trans, because it will be turned on, and allowed to settle down, before the DUT is turned on.
Ok man, this is a fantastic idea, but when you push hard the amp, your rail go down and this can cause the relay's disengaging. Probably the best way is a timer to set a treshold time to engage the relay and if you want, with other components, create a sort of protection at start up if there are problems in the amp stages!
In the video, I show that the relay drops out when the caps are down to about 6 volts. If the power supply drops to 6 volts during normal use, the relay will be the least of our problems! ;)
excellent video really helpful! could someone maybe help me out a bit? 1) the wiring to ground from the relay coil as depicted in the schematic drawing; is it wired to B-? my amp's got a split rail voltage, there's 46V difference between B+ and B-, and a ground in between (23/23). i reckon i CANNOT just go wire the relay coil to the mains ground... where *do* i wire it to? 2) i live in Europe, 230Vac on the line, shouldn't the big resister be rated, like, over 1kW in power dissipation? thanks in advance, all
Wouldn't it bei easier to parallel the relay to the speaker protection relay, at least If it's circuit can provide enough power and it kicks in even with the reduced current?
IF an amp has a lower voltage winding and DC power for items such as the SP relay, or meters, you could use that power for the relay coil, as long as it didn't overload/undervolt any regulating components feeding it. There could be some future "troubleshooting" issues arising from interlinking 2 protection circuits like that though, especially because the SS add on won't be on a repair schematic, and may not be obvious to a tech that wasn't aware of what it was there for. Personally, I don't think i'd draw power away from the SP relay circuit, but that would be my choice.
Came across this video as it randomly popped up….whilst looking for a way to stop my 3hp router from throwing me over the bench every time I switch it on….. way over my head but very interesting all the same.!! Why was there no current limiter built in from the factory you ask.? Because like every manufacturer of every machine be it machanical, electrical or both, reliabillity, longevity and reputation all come a poor second to repeat business and profitability..!!!!👍
preferring transistor than relay for realizing smoother and safer soft start function, and hoping the circuit diagram fixed on the table and viewed by your camera with smooth panning and enlarging in order for watcher to view the circuit exactly, best...
you know the resistance of the coil, you know the voltage drop range across the coil, so you know the current. knowing the current and knowing the total voltage drop across the coil + R1, you calculate it as R1=(Utotal/I) - Rcoil. Very basic application of the Ohm law.
but as many have said, there are other ways to do it, some of which may be considered better. For example, I don't like the relay, since it's a mechanical part prone to wear and in some cases contact burning wear (which may be a non-issue in this application though). will something like a mosfet with a gradually increasing Vgs (set, via a resistor divider, by the capacitor charge voltage) and respectively decreasing Rds down to a few mOhms work? maybe, need to see if it copes with the power that it has to dissipate while the load is powering on.
@@s_s-g4d That, and if you design it with a "current" (pardon the pun) model of production, you can replace it fairly easily, and , maybe even buy a spare and secure it inside of the unit for future use. It's unfortunate that a component line's production has a limited, but unknown lifespan, but I guess that is some sort of "progress".........or is it?
@@xraytonyb Yes, but if the chassis becomes live due to a fault on the primary side or the transformer fails, the mains supply protection won't detect it and trip. Only if the whole unit is double insulated, class 2 is an ungrounded unit permitted in the regs/code, at least in the UK.
I love this and many other technical channels from the USA, but I am always surprised by how many metal chassis items of this era have no safety ground connection - or appear to have none of the internal design that would make it permissible for class 2 (no ground) operation here in the UK - even for equipment of the age of this Marantz.
Yes…. But it robs a volt or two which reduces power the amplifier can deliver. Also it is no good with a quick on/ off sequence since it needs time to cool down to gain resistance. Ie it’s no more a soft start unless it can cool down
When you say, “you can add fly back diodes if you want, it won’t hurt anything,” I’ve read (and can dig up a hard copy of) an article by a relay manufacturer that said that reflexive use of fly back diodes can and do harm the contacts of relays. The manufacturer was objecting to the very thinking you are commenting on, that is, that many feel that if the relay is DC, they need to include a fly back diode. I guess what I’m trying to say (and not doing a very good job of it) is that while in the case the circuit presented, i.e., the soft start circuit, the use of fly back diodes won’t hurt anything; it is not true in every case that the use of fly back diodes won’t hurt anything. If I recall correctly, I think the relay manufacturer was advising the addition of a resistor in series with the fly back diode to cut down on the rush of ‘fly back current and resulting arc flash destroying the relay contacts. I need to go back and re-read the article. As others have stated, I found the entire video very interesting and informative. I have a tube amp with soft start. The manual is at pains to say never to turn the amp off and then immediately back on or damage to the power supply may result. Because I managed to inadvertently perform that maneuver a few times, I put a label on top “1 Minute” to remind myself. I’m not sure how their soft start circuit is constructed, but you analysis helps me understand better how and why such a delay is recommended. I do like how you break everything down, your passion is contagious. Going to dig that article out now.
An extremely simple (and low cost} soft start "circuit" consists of a single thermistor which adds resistance when cold and becomes more conductive as it heats up. Without a relay to bypass the thermistor it will remain hot when the amp is powered, and after power is disconnected it will take some time for it to cool down. Until the thermistor can cool down sufficiently it will not be an effective current limiter. A rapid off-on cycle defeats it.
Wikipedia has an article called "Flyback diode". The folks who wrote that article explain quite well the subject, including a screenshot of the transient negative voltage spike across the coil of the relay. In summary, the electromagnetic energy stored in the coil needs to be dissipated (or transformed) immediately when the current across the coil is turned off (since V = L di/dt). With a flyback diode installed, the energy will be dissipated as heat on the resistance of the coil in a fraction of a second when the relay is turned off. If that coil resistance is not appropriate then a resistor needs to be installed in series with the flyback diode. I have never seen ordinary relays that come with a flyback diode installed at the factory (but I am sure that they exist), but I have seen and used reed switches that can come with a flyback diode already installed for the exact same purpose (see for example item number HE3621A0500 at Mouser or DigiKey -- "Optional coil suppression diode to protect coil drive circuits"). Related to your point, that Wikipedia article says: "Design: When used with a DC coil relay, a flyback diode can cause delayed drop-out of the contacts when power is removed, due to the continued circulation of current in the relay coil and diode. When rapid opening of the contacts is important, a resistor or reverse-biased Zener diode can be placed in series with the diode to help dissipate the coil energy faster, at the expense of higher voltage at the switch..."
@@kylesmithiii6150 That’s great and very helpful, as I couldn’t put my hands the article I cited even though I saw it recently. Thank you not only for the reference to Wikipedia, but your cogent interpretation to the matter at hand. Like you, I’ve never seen a relay from the manufacturer with a fly back diode installed, I seen many DC relays installed with fly back diodes in the field.
@@steveboguslawski114 It seems to me that this circuit has a problem similar to the thermister circuit. The rail voltage has to drop quite a bit before the relay opens.
I have the Atlas ESR Gold that claims that it can measure up to 90,000uF, but so far I did not try measuring a capacitor larger than 1,000uF. My Siglent SDM3045X can only go up to 10,000uF and again I never tried over 1,000uF.
This is becoming The Capacitor Channel. I wish you’d focus more on other components. Otherwise, it’s becoming like all the others that do little more than shotgun capacitors. It’s really boring.
just wanted to say I think its pretty dang cool that you take the time to make these videos and share your knowledge and personal perspective with those of us floundering through the forest in search of lupins..
Yet another educational and entertaining piece of work that you did Tony. Many teachers could learn from you how to combine these two elements when explaining a subject, whatever that subject is, not only electronics. What I personally find very useful in your videos is a logical structure behind what you present. Starting with explanation of the idea and theoretical background, then diving into a practical application and finally demonstrating certain number of tests that not only prove what was said before, but also helps to better understand that theory. We know you don't use script. You simply don't need it. Excellent job.
Such a simple circuit for such an important job!👏👏👏👍
A much more elegant and easy solution for inrush, is to use a NTC thermistor device on the line. We utilized this method on all of our high current amps at Altec-Lansing, which included the EV line as well. - K5DBX
Excellent! I just subscribed after seeing this elegant solution. I build tube amps and I typically heat the early stage filaments with DC to limit AC hum. To do this you need a lot of capacitance because 4 tubes can draw a couple amps and you need that capacitance to keep up with the ripple. But this does present a big inrush strain on the filament winding of the transformer. For 6,3 volt filaments maybe I'll experiment with a 5V or 6V relay. If I can find a sweet spot, I'll definitely add this to all my projects. Many people use a 555 timer or such, but this is downright simple with far fewer points of failure. Thanks!
Thank you for the added information displayed in yellow text. That is always helpful - especially when dealing with numbers, values, and equations. I usually take a screenshot of that so that I can study it at the end of the video. Details matter. 👍
Hi Tony. If you have a relay with an extra set of contacts (DPDT), you can wire in an extra resistor (whose value will have to be determined by experimentation) in series with the one already there, and put this resistor in parallel with the normally-closed contacts so the resistor is shorted on initial power-up. Once the relay pulls in, the N.C. contacts open, and the 2nd resistor is now in the circuit so the relay only needs the hold-in voltage/current. This should extend the life of the relay coil. You may need a small capacitor to prevent relay chatter, though, across the N.C. contacts/resistor. Maybe .01-.1 uF disk capacitor.
HI Tony, that is an elegant way to do the job and will easily work on anything.
Rich
24:54 If you want, you can add a small neon bulb in parallel with the relay coil. During normal operation, the breakdown voltage of the neon bulb is way higher then the voltage of the protection circuit itself and the neon bulb acts like a open circuit. When back EMF is generated by the relay coil, the voltage will be clamped down once it reach the breakdown voltage of the neon bulb.
You could also use a diac, which is like two 30 volt zener diodes in reverse-series.
I made a similar in-rush limiter on my DIY variable DC power supply. The input voltage of the supply is 50V DC, rectified from a 36VAC toroidal transformer. I put a 330ohm 5W resistor in series with the 4x4700uF parallel caps, so when powered on, they charge slowly. The relay coil (48V) is connected directly to the caps, the voltage increases slowly as the caps charge up thru the 330 ohms, and when they reach around 38-40V, the relay pulls in and cancels the resistor. Works well, takes around 3 seconds from a cold start. 48V relay coil has no objections to 50V so I did not use a dropper resistor there. Now I see this doesn't help with the in-rush of the transformer, so I need to re-think the whole thing. Thanks for the video.
The only difference I see is that you put the current limiting resistor (330 ohm) on the output of the transformer instead of the input. I'm guessing that you've accomplished the same thing, in that you are limiting the inrush current (to the caps) that the transformer secondary is supplying at a slower rate. Seems good to me. 🤷♂
That a great circuit. I do a lot of power supply's and Tube RF amps that have large capacitors . Them values of inrush resistors can change with size of transformer and voltage of line. with that amp in video you could use 48V relay coil. 24 DC is very common. Also that inrush current is hard on diodes . I had so many go bad . even broadcast transmitters use soft start . just 3 resistors for 3 phase with a contactor across. and the output tubes filament supply has step start . taps on primary of filament transformer. There is a outher soft start circuit that is similar and works great I use on tube RF amps. ham radio type. using a 240V or 120V relay coil across transformer primary. then the series resistor is 20 to 30 ohm 240V 10 ohm 120V 50 watt or so with a 2 or 3 amp slow blow fuse in series with resistor. If you have a diode short or tube short the 2 or 3 amp slow blow fuse blows saving the big surge. I copied the soft start from 3 phase and made it for smaller units. mind you running 480V with 277V to natural. 277V contactor. 500 watt resistors in a cage . 3 cartage fuses 600V . does same thing. large or small that soft start really works. Thank you,😀
I would have put the resistor in series with the B+ and use a P-channel MOSFET wired across maybe a 10 ohm resistor, but either way it's a good idea.
Thanks a lot. I am planning to build a power amp using a 625W, 30v 20A (x2) toroidal transformer and 2 large capacitors (22000uF). I was considering using an Inrush limiting NTC thermistor but I realized that during quick on/off cycles may not work well. I also realized that some high end vintage amplifiers (like Nakamichi 620) are using relays as you are suggesting. Unless somebody has a good reason to use Inrush limiting NTC thermistor, I will go with your suggestion. Thanks again.
I like this solution better than the thermal ones. Quick cycling can defeat those. Whether it's someone playing with the power switch or just quick power-outs, it's bound to happen.
The thing that bothers me is that the rail voltage goes pretty low before the relay opens. I'd rather have it open when the capacitors drop to 80% charged.
PeaceJoyHappines, thanks for video(s) Tony. _Morning hours across pond(scandinavia) here, at about 08am ..Coffe & XTB-video ..ThumbsUp!
You could also use a Infineon IRF3205PbF (RDS(on) = 8.0mΩ) power mosfet across the 4R7, rated for 110A @ 55V on the secondary side of the transformer with a 10K / 10R voltage divider on the capacitor.
No power resistors needed and no useless heat generated, simulator shows it works fine, it gradually start opening the mosfet.
Hell yea Tony! UA-cam was lacking a good video on this subject.
You make very good instructional videos as your explanations are very thorough.
Regarding the Marantz design, I am surprised that 60Volt caps are being used with approximately 50 volts applied across them. 50/60 would be an 83% stress level on those caps, and if it's 57 volts, as you stated later in the video, that would represent a 95% stress level. Could be why one of them failed open. Mil Spec defines high reliability, at stress levels of 50%. I'd imagine the form factor of the caps, available at the time, was the major consideration, but newer caps have smaller form factors and if it were my amp, I'd try to get 100 Volt caps.
Also not to nit pick, but you didn't mention that the inrush current also stresses the caps, you mentioned that it could hurt the power switch and x-fmr, (you probably just forgot to mention). Just an observation, not meaning to be critical for the sake of being critical, just providing some feedback. All in all I enjoy your videos, and find them very practical. Thanks for taking the time for the tutorials, as they are probably very time consuming and difficult to pull off.
That big inrush can damage all in its path. A 6 volt relay has a 30 ohm coil if your initial charge is through the relays coil when the difference in voltage is less than 2,5 volts the relay drops out and the Cap is directly connected to the source through NC contacts
Current limiting is essential .the inrush current will damage both the transformer and the capacitors .And since audio quality can be greatly affected by power supply it makes sense to protect your power supply and your equipment connected to it
I thought you would like to know that I’m going to put this soft start circuit you show into an amplifier I am recapping for a guy. I don’t support this necessarily but he has a Hafler DH-120 power amp and he bought o capacitor upgrade kit for it which among other things increases the main filter caps from 6,800uF to 22,000uF. It seems to be a common thing for owners of this amp to do. I advised him not to do it but recommended that if he really wanted it done, he should at the very least add the soft start circuit and upgrade the rectifier. I still couldn’t take responsibility for what this modification might do. I don’t recommend it but it’s his call. Without looking at a schematic, other than start up, would you have any other potential concerns with caps this large put into a circuit that wasn’t designed for it?
Anyway, I just wanted to let you know that the videos you put out are a great educational tool for people like me. You are a priceless asset for the community. Thank You
Great circuit addition, however I like to be more conservative and put a capacitor across the relay coil to increase the delay time, and increase the power resistor to limit the in rush current more. I have found that this is less stressful on the rectifiers and main filter capacitors and increases their life.
The purpose of the 2k resistor is to set the level at which the relay comes in. As it is, the relay won't come in until the capacitor bank is charged above 75%. Adjusting the value of the resistor adjusts the trip level of the relay. Adding another capacitor across the relay is unnecessary.
I'd probably just wire a NTC thermistor in line with the mains winding, if I could find anything durable enough for the job. When turning the amp on, this thing's resistance is a few ohms, decreasing as it heats up under current. It will eventually reach a steady state temperature and resistance, dissipating some heat.
BTW, the dim bulb is a PTC thermistor.
I also use a NTC thermistor
I do the same thing. I add one in every receiver I restore. I'd be interested if anyone had reasons why using an NTC thermistor as an inrush current limiter is a bad idea.
@@davewhittaker9547 I'm not sure if anyone would argue it's a BAD idea. The repeat power-on problem is probably the most valid reason to not use them. If you've got balance/bleeder resistors across the caps- or any parallel resistance to 0V for that matter- the caps will likely drain way before the NTC cools down to its resistive state after powering down. So powering on the device again would theoretically expose it to inrush current conditions despite having that NTC in there. I know some people might not like them because of their non-linear resistive qualities at operating temp. They'll often bypass them with a relay to take 'em out of circuit once they've done their job. That's kind of a counterintuitive use of the NTC technology in my opinion. If you're going to involve relays and bypass the damn thing then why not use a resistor like Tony's doing here. NTC's are fine, I use them sometimes if space is an issue. But if you're anticipating a problem and plan on doing something about it then you might as well do the best you can within your means. Woa I said a lot of words. 😅
@@deviantmultimedia9497 I see what you're saying. I never thought about the recovery time, but it could be an issue. Not often, but every once in a while I might quickly power cycle it. I need to give this some thought; especially since I also routinely increase the size of the filter/power capacitors near the transformer. I'm curious now, and I think I'll do some experiments on how long the NTC thermistors I use take to recover. Thanks for taking the time to post the explanation.
A straight forward & simple circuit.
I.M.H.O. wiring with relay across the entire dc
(+/- ....& not +/ ground ) with appropriate resistance and suitably rated zener....this ensures we see sufficient voltage buildup on " both " caps.. instead of only one..
( eg for +/-24 dc , a 22/27V zener) .. in series with the resistance and relay coil...the coil must have its f.w.d. across it...
Relays can pick up @70% rated volts and will drop out when voltage falls below 30% rated volts.....A24 v relay,once energised... will reliabily remain energized between approx. 17 & 9 volts......
I got to say my halfler DH 200 is a heavy hitter my lights dim turn it on so I just leave it plugged in and leave it on all times it is a nice and warm at all times
Even without caps, the inrush is huge because of that transformer.
Great video. I'll be applying this at some point.
The min max button on the Fluke would be useful to capture that full momentary voltage drop.
Thanks really appreciate your tweak.
I'm wondering if an inrush current limiter (like Amphenol CL series) would be an adequate alternative solution. I know that it's a tiny space heater, which is not ideal, but with so much empty space inside it shouldn't be a problem.
Oh, and thank you so much for sharing your wisdom, yet again! Please know, it's appreciated beyond words.
I was thinking about a fly back diode may increase the time of the kick out when turning the power off. Isn’t the caps taking the charge built from the magnetic field of the coil? Curious if the relay is actually causing the capacitors to bleed out it’s charge. Am I asking right?
Poor man's soft start is an NTC thermistor.
And cheaper too 👍
Simple & cheap. Will put its finite hot state resistance there it's the only drawback. Can modulate mains draw a bit along with bass transients if the amp is really pushed. Would be a non issue to a pure class a amp with same power draw at any output level but such beasts are rare.
Nice. Thanks to you i just discovered that my amp has something similar already installed ( Onkyo M 509 ). But it just uses 5*0.22r5w and no resistor before the coil but a cap and a diode across the coil. Now im thinking about changing the 5*0.22r ones against one 4r50w because it has new and bigger filtercaps now too ;) (from 4*30000 to 33000) Oh and its one relais for both transformers.
Could you replace the 4 ohm resistor with an NTC and then instead of using just a resistor inline with the relay coil, add a zener diode. The zener will give better control of the relay switching point and increase the drop-out voltage.
Both of those suggestions sound workable, to me anyways. Would be fun to breadboard up and see exactly how things would behave.
How many amps is the secondary windings producing? And what is the DC voltage after your transformer is rectified? +58-0-58 +60-0-60 DCV ? Thanks for sharing the video great insight much appreciated and I wish you a 🦃 Happy Thanksgiving💯🤗
The 510 power uses the relay with primary only and needs a cap across the relay’s coil
Pulls in almost instantly
I've done similar using a scr that slowly (over 10 cucles shorts out a resistor that is in series with the input of the transistor. The only thing agaisnt this is even wen it's full on it will still rob a volt of the incoming votge because the acr does not go to zero.
Couldn't a zener be connected in series with the relay coil instead of a resistor to make the threshold more decisive?
It could. I'd actually recommend still using the resistor, but adding a voltage threshold detector circuit (a Zener and a BJT)
That is a good idea, but the EMF voltage from the relay coil when the coil is abruptly turned off can be easily over 100 Volts and this would instantly destroy the Zener diode(s) installed between the 33000 uF cap and the relay coil. If you do install a flyback diode in parallel to the relay coil, that will mitigate that possibility. I think that the circuit presented in this video had a purpose of being simple, effective and durable.
@@kylesmithiii6150 Why would it? The zener would clamp the voltage.
@@kylesmithiii6150 Suitably rated Zener diode in series with the 2k/5W resistor...will give an additional hold-off voltage & time delay.. before relay kicks in ..a F.W.D. across coil is a must.
@@NiHaoMike64 Since the voltage on an inductor is V = L di/dt , so if the current across the coil changes rapidly then there will be a high voltage on it (actually a reverse voltage in this case since di/dt is negative). In this case, because the relay coil is in series with a large RC circuit, and C is very large (33000 uF), then di/dt should not be big at all. It can be seen on an oscilloscope. At 27m 57s Tony showed the peak voltage on the 33000 uF, it is about +57 VDC. Let’s suppose that the capacitor discharges rapidly when the amp is turned off and the EMF voltage of the relay coil is 10 V peak, so the 2k (5W) resistor will take the +57 - (-10) = 67 V peak and dissipate the energy rapidly on itself. All will be fine. If you put a Zener diode in series with the 2k resistor (and I do like the idea!), say a 12 V Zener to set the threshold, then first you will have to recalculate the value of the resistor = (57-12-24) / 30 mA = 700 Ohms (2W). Zener value = 12 V (1W) or use 2 x 6 V Zeners in series (0.5W each). You will have to see on the oscilloscope the EMF of the relay coil to see if the Zener(s) can take it, or just install a flyback diode in parallel to the coil and it should be fine. (I am adopting the nominal current of the relay to be 24 V / 840 Ohms = 30 mA) (Tony measured the resistance of the coil to be 840 Ohms - see clip at 15m 56s)
Thanks for your point of view Professor. Enjoyed your session.
Very good!
The idea is good, but I would strongly recommend to use a solid state relay.
Hello! You always have such great videos! I am learning a lot! Concerning the control of the capacitors and their charging, why not just add a 30 ohm, 100watt resistor on the positive side on each capacitor to limit the inrush of current when powered on?
they need not to have any series resistance after they are initially charged, that's why the resistor has to be shorted after they charge.
A good, practical build for a soft start, thanks for sharing it! One thing I wonder about relays that are "soft started", such as ones used for your circuit here or for DC protection on amp outputs, is, does the "slow" build up of DC voltage on the coil to close the contacts make a difference over time to any "self cleaning" properties that a hard "0 to working voltage" application may cause. I'm wondering that, if contacts close a bit harder, will they "self clean" any better.
great question. My gut feeling isit won't make any significant difference in the contact cleaning as the pent up tension in the spring provides most of the cleaning power.
Tony, what is the current rating of the relay contacts? That needs to be considered also. Very good video, thanks.
Very cool.
How long is time delay before the relay "kicks in" without the dim-bulb ? From what I've read , you need at least a 250ms relay turn on delay. In addition the larger the trans output voltage and caps on the supply the greater the "turn on delay" should be. I have about 1 sec delay with an effective bleed resistance of about 50 ohms (4 * 200 ohm at 5 watts ea), but the soft start circuit is more complex (complete with a relay turn-on LED) in its own small case.
In this setup the delay already depends on the time it takes for the circuit to reach the preset voltage.
Can I use soft start from (or designed for) power tools? They seem to be cheap and made in small nice packages easy to fit to audio amplifier. Not sure if I should be trying such devices for audio amplifiers.
Tony, I am curious about your choice of using the 2k ohm resistor to drop the 57 volts to 24 volts for the relay. If one plugs those in a voltage divider calculator with 850 ohms for the relay the voltage across the relay is 17 volts. I’m trying to figure out what dropping resistor I will need to use with the very same relay with a B+ of 70 volts. Using a voltage divider circuit calculator, it comes up with a 1667 ohm resistor.
I thoroughly enjoy watching your videos
Larry
A little off topic, but how about using a soft start with a SCR voltage regulator, a 100A rated bridge rectifier and 10,000uF filtering caps driving a DC motor?
Wonder if it's a good idea to soft-start a decently sized SMPS like this. Proper ones will have an NTC on the input, but if that's already hot after it's been in use for a while, it can very well act like a piece of wire, passing full current to the main cap(s) and causing arcing across switches/relays. I WAS thinking of using this method, but I heard it's not recommended to "starve" a SMPS of current for too long by placing a resistor/light bulb in series with its mains input, so I guess the resistor value and "hold" time would have to be pretty precise so as not to reduce its lifespan. If it's got a PFC boost circuit it's probably even worse...
excellent work, quite educational.
does the loud 'thump' some amps produce when plugged in or coming out of sleep is also the effect of a large inrush current?
my bass guitar amp makes one when turned ON and also OFF. as i recall the power caps are way smaller though, i think 2 caps @5600uF.
i could try powering it thru a light bulb like shown here.. :}
The thump is because the amp does not have a speaker protect relay. Until the transistors come up to their proper bias level, they could have uneven voltages on the outputs, which causes a temporary dc offset.
@@xraytonyb thank you for the reply, that makes sense
To keep it simple how about just adding a CL-30 or 40 or other properly sized thermistor after the fuse ? They are ideal for limiting inrush current and with today's higher line voltages they might drop that a volt or 2 after warm up.
I’m afraid this won’t work in Europe or anywhere else where the grid AC is 230V. The inrush resistor’s power capacity will have to be prohibitively high; we’re talking quadruple-digit wattages here.
I'd implement a thyristor-based soft start circuit in that case.
learnt a lot...cheers.
Tony, Great information. I repair welding equipment for my living and all of the welding inverter power supplies I repair employ a similar "pre-charge" circuit for the buss caps. I deal with caps at least as big the two here. How do you measure them? I cannot find a meter over 10k. Thank you so much!
Another nice video and explanation but since I don't have a test setup, how much does that 4 ohm resistor reduce the inrush current? And more out of curiosity, how much is 4 ohms (percentage wise) in relation to the power x-fmr primary resistance?
Wanted or not, I will throw in some feedback. (and let the arguments begin 🤣🤣🤣) Believe it or not, you posed kind of a "trick question" , because, since the line voltage is "alternating current" it depends on the exact timing on when the power switch makes contact in relation to the AC cycle. If you switch it on during the "zero crossing" part of the cycle, it would have to reduce less power than if you switched it on during the peak. That, and adding any residual magnetism stored in the transformers body, and also, when the switch energizes it, things get crazy. If you convert things to DC though, you can do some "fixed value" math, and lets say ( ballpark-worst case scenario, an amp that keeps tripping a circuit breaker when you power it on) you have a 15A surge, which, may trip a standard US home branch circuit breaker, at 120V, you would be drawing 1800 watts for a millisecond or so. Ohms law works that out to an 8 ohm load. Adding 4 ohms to that makes it 12 ohms, dropping that current to 10 amps from 15, which is a bit easier on the switch contact, but keep in mind that is worst ( guestimated) case scenario with 60 cycle AC. And, also, keep in mind, that the transformers "resistance", once connected to AC line power, is actually mostly inductive impedance, so, although it may have a lower DC resistance, once it is connected to the AC frequency it was designed for, it will look like it is a higher resistive value than its DC measurement is. That impedance value depends totally on the differing transformers designs, so it will have no set answer as well.
I would be interested in a similar circuit that uses an Solid state device in place of relay. Maybe SCR
www.analog.com/media/en/technical-documentation/data-sheets/LTC7004.pdf
Would a PTC thermistor used as an inrush current limiter work in that?
PTC thermistors increase resistance with increase in temperature. You would want to use an NTC. The issue with them is that they get pretty hot when they are at their low resistance level. This adds a lot of heat to the inside of the amp. That's why i don't like to use them.
@@xraytonyb ... and since they run hot, they present resistance, which increases output impedance of the power supply - not a welcome addiition in an amplifier, unless you're looking to limit its instantaneous current delivery capability.
What about an NTC thermistor like it's is used in switch mode power supplies, maybe more efficient.
Might your soft start circuit be even more simplified, by using a Mains VAC Supplied STDP Relay? The resistor is wired across the Normally-Open contacts of the Relay. The Line side feed goes through the resistor, to limit the in-rush to the relay and the load. Once the relay and load reach 80% of the mains voltage (80% being the typical voltage at which a relay engages,) the relay engages, thereby bypassing the resistor, and the load finishes charging (but much less violently.) Not my idea = Credit goes to james_s, on the EEVBlog Forum. Thanks, james_s for (originally) suggesting this to me.
Hi Tony…..I’m wondering why you put the 4 ohm resistor on the mains instead of on the output of the transformer. Wouldn’t it operate the same way and still protect the transformer? Just curious. I’m adding one now and would like to know. Thank You
My thought is that relay contacts are rated 2 to 4 times higher on AC current than DC current
@@jimnunn9232 The secondary of the transformer is still AC, just lower voltage. Was thinking that might be better
Tony,
What is the voltage drop across that 4.7 ohms resistor?. Is that enough to have that same voltage drop in parallel with the relay coil, so that the relay turns off after the capacitors are charged? This way the relay will not need to stay on all the time when the amplifier is on. Would this be a suitable alternative way to do this?
Will the soft start save the speakers, especially the tweeters?
As the amp in question has a split rail supply did you put the resistor on the 0v/ground rail? Thanks for the upload.
Why is a relay preferred over a MOSFET or SCR in inrush limiters?
Hi, Tony. A related question, please...
I am building my DIY AC PSU chain. It consists of:
1) Emergency All Stop Button to
2) Isolation Transformer to
3) Variable Autotransformer to
4) Dim Bulb Current Limiter (Multi-Bulb) to
5) DUT.
Each piece is independent, so that it can be used separately, if need be, for other purposes.
Might it be helpful to include a NTC Current Limiter? Advantages/Disadvantages? If so, which component should I add it to? The input on the Dim Bulb Current Limiter? It sort of seems redundant, unless all the bulbs are bypassed... But, I won't always be using the Dim Bulb, in the chain.
Thank you for your help.I posted
Disclosure> I posted this question on a forum, but I never got a response.
Got roughly the same idea for a mains testing box in my lab, only integrated and not separated.
NTC is useful before transformers, especially toroidal. These things can have an inrush current on their own.
@@KeritechElectronics Thanks, Keri. The Isolation Transformer is the problem, it keeps popping the low amp fuse that is in the Emergency Stop Button. It makes sense to add the NTC to the Iso Trans, because it will be turned on, and allowed to settle down, before the DUT is turned on.
The large transformer will create a large in-rush current by itself even without capacitors.
Question is a tungsten long tube bulb a good substitute for an incandescent for a Dim Bulb current limiter ??
WOT? A halogen lamp *IS* an incandescent lamp.
Hello, where can i contact you about the product sponsorship?
Ok man, this is a fantastic idea, but when you push hard the amp, your rail go down and this can cause the relay's disengaging.
Probably the best way is a timer to set a treshold time to engage the relay and if you want, with other components, create a sort of protection at start up if there are problems in the amp stages!
I highly doubt the rail voltage would dip down low enough to disengage the soft start relay.
In the video, I show that the relay drops out when the caps are down to about 6 volts. If the power supply drops to 6 volts during normal use, the relay will be the least of our problems! ;)
excellent video really helpful!
could someone maybe help me out a bit?
1) the wiring to ground from the relay coil as depicted in the schematic drawing; is it wired to B-? my amp's got a split rail voltage, there's 46V difference between B+ and B-, and a ground in between (23/23). i reckon i CANNOT just go wire the relay coil to the mains ground... where *do* i wire it to?
2) i live in Europe, 230Vac on the line, shouldn't the big resister be rated, like, over 1kW in power dissipation?
thanks in advance, all
Your bench looks like mine.
Wouldn't it bei easier to parallel the relay to the speaker protection relay, at least If it's circuit can provide enough power and it kicks in even with the reduced current?
IF an amp has a lower voltage winding and DC power for items such as the SP relay, or meters, you could use that power for the relay coil, as long as it didn't overload/undervolt any regulating components feeding it. There could be some future "troubleshooting" issues arising from interlinking 2 protection circuits like that though, especially because the SS add on won't be on a repair schematic, and may not be obvious to a tech that wasn't aware of what it was there for. Personally, I don't think i'd draw power away from the SP relay circuit, but that would be my choice.
Came across this video as it randomly popped up….whilst looking for a way to stop my 3hp router from throwing me over the bench every time I switch it on….. way over my head but very interesting all the same.!! Why was there no current limiter built in from the factory you ask.? Because like every manufacturer of every machine be it machanical, electrical or both, reliabillity, longevity and reputation all come a poor second to repeat business and profitability..!!!!👍
preferring transistor than relay for realizing smoother and safer soft start function, and hoping the circuit diagram fixed on the table and viewed by your camera with smooth panning and enlarging in order for watcher to view the circuit exactly, best...
Could you use a varristor?
I believe I addressed that question somewhere in the video.
isn't it better to use like 100 w SMD resistor smaller should get less hot, imagine that gets quite hot that 4 ohm resistor??
Todderbert i like your utube videos
But how to calculate value of R1? How can we know what current will flow through it?
P.S. Can't find current value for my relay....
you know the resistance of the coil, you know the voltage drop range across the coil, so you know the current. knowing the current and knowing the total voltage drop across the coil + R1, you calculate it as R1=(Utotal/I) - Rcoil. Very basic application of the Ohm law.
but as many have said, there are other ways to do it, some of which may be considered better. For example, I don't like the relay, since it's a mechanical part prone to wear and in some cases contact burning wear (which may be a non-issue in this application though). will something like a mosfet with a gradually increasing Vgs (set, via a resistor divider, by the capacitor charge voltage) and respectively decreasing Rds down to a few mOhms work? maybe, need to see if it copes with the power that it has to dissipate while the load is powering on.
the relay method, however, is bulletproof and as simple as it gets, which is definitely one of its strong sides.
@@s_s-g4d That, and if you design it with a "current" (pardon the pun) model of production, you can replace it fairly easily, and , maybe even buy a spare and secure it inside of the unit for future use. It's unfortunate that a component line's production has a limited, but unknown lifespan, but I guess that is some sort of "progress".........or is it?
Want to comment on the use of thermistors for the same reason? Pros and cons?
I did on the video following this one
The amplifier doesn't appear to have a mains ground connection, is it double insulated?
The transformer provides isolation.
@@xraytonyb Yes, but if the chassis becomes live due to a fault on the primary side or the transformer fails, the mains supply protection won't detect it and trip. Only if the whole unit is double insulated, class 2 is an ungrounded unit permitted in the regs/code, at least in the UK.
I love this and many other technical channels from the USA, but I am always surprised by how many metal chassis items of this era have no safety ground connection - or appear to have none of the internal design that would make it permissible for class 2 (no ground) operation here in the UK - even for equipment of the age of this Marantz.
Wouldn't a NTC thermistor be much less complicated and more reliable? Fewer components, no moving parts.
Too much heat, in a already cramped little amp
Yes…. But it robs a volt or two which reduces power the amplifier can deliver. Also it is no good with a quick on/ off sequence since it needs time to cool down to gain resistance. Ie it’s no more a soft start unless it can cool down
@@Starcraftowns99 This is true!
When you say, “you can add fly back diodes if you want, it won’t hurt anything,” I’ve read (and can dig up a hard copy of) an article by a relay manufacturer that said that reflexive use of fly back diodes can and do harm the contacts of relays. The manufacturer was objecting to the very thinking you are commenting on, that is, that many feel that if the relay is DC, they need to include a fly back diode. I guess what I’m trying to say (and not doing a very good job of it) is that while in the case the circuit presented, i.e., the soft start circuit, the use of fly back diodes won’t hurt anything; it is not true in every case that the use of fly back diodes won’t hurt anything. If I recall correctly, I think the relay manufacturer was advising the addition of a resistor in series with the fly back diode to cut down on the rush of ‘fly back current and resulting arc flash destroying the relay contacts. I need to go back and re-read the article.
As others have stated, I found the entire video very interesting and informative. I have a tube amp with soft start. The manual is at pains to say never to turn the amp off and then immediately back on or damage to the power supply may result. Because I managed to inadvertently perform that maneuver a few times, I put a label on top “1 Minute” to remind myself. I’m not sure how their soft start circuit is constructed, but you analysis helps me understand better how and why such a delay is recommended. I do like how you break everything down, your passion is contagious. Going to dig that article out now.
An extremely simple (and low cost} soft start "circuit" consists of a single thermistor which adds resistance when cold and becomes more conductive as it heats up. Without a relay to bypass the thermistor it will remain hot when the amp is powered, and after power is disconnected it will take some time for it to cool down. Until the thermistor can cool down sufficiently it will not be an effective current limiter. A rapid off-on cycle defeats it.
@@steveboguslawski114 Thank you for the response; you’ve increased my understanding of why I need to wait. The manual was not very technical.
Wikipedia has an article called "Flyback diode". The folks who wrote that article explain quite well the subject, including a screenshot of the transient negative voltage spike across the coil of the relay. In summary, the electromagnetic energy stored in the coil needs to be dissipated (or transformed) immediately when the current across the coil is turned off (since V = L di/dt). With a flyback diode installed, the energy will be dissipated as heat on the resistance of the coil in a fraction of a second when the relay is turned off. If that coil resistance is not appropriate then a resistor needs to be installed in series with the flyback diode. I have never seen ordinary relays that come with a flyback diode installed at the factory (but I am sure that they exist), but I have seen and used reed switches that can come with a flyback diode already installed for the exact same purpose (see for example item number HE3621A0500 at Mouser or DigiKey -- "Optional coil suppression diode to protect coil drive circuits"). Related to your point, that Wikipedia article says: "Design: When used with a DC coil relay, a flyback diode can cause delayed drop-out of the contacts when power is removed, due to the continued circulation of current in the relay coil and diode. When rapid opening of the contacts is important, a resistor or reverse-biased Zener diode can be placed in series with the diode to help dissipate the coil energy faster, at the expense of higher voltage at the switch..."
@@kylesmithiii6150 That’s great and very helpful, as I couldn’t put my hands the article I cited even though I saw it recently. Thank you not only for the reference to Wikipedia, but your cogent interpretation to the matter at hand. Like you, I’ve never seen a relay from the manufacturer with a fly back diode installed, I seen many DC relays installed with fly back diodes in the field.
@@steveboguslawski114 It seems to me that this circuit has a problem similar to the thermister circuit. The rail voltage has to drop quite a bit before the relay opens.
I have the Atlas ESR Gold that claims that it can measure up to 90,000uF, but so far I did not try measuring a capacitor larger than 1,000uF. My Siglent SDM3045X can only go up to 10,000uF and again I never tried over 1,000uF.
You should NEVER use the "aux" AC outlets on a pre-amp to feed a Power Amp. The Phase Linear 400 didn't have a power switch either.
😂😂😂 meteor shower
👍🙏
Noooo...a poor mans soft start is just a Dim Bulb setup thats easy, simple ) problem is incandescent bulbs are banned in EU at least in favourbof LEDs
You can buy incandescent bulbs easily. You can find them in most people's cupboards.
Math's not math.
Not math's either. That implies math owns something.
@@sw6188 Doing 'sum's'
@@andymouse It amazes me the number of people who struggle with apostrophes. They're not that hard.
@@sw6188 "it's a very cold day to walk the dog. Let's put on its sweater." 🤦♂😄
Спасибо! Было интересно 😊
This is becoming The Capacitor Channel. I wish you’d focus more on other components. Otherwise, it’s becoming like all the others that do little more than shotgun capacitors. It’s really boring.
If you find the channel boring, don't watch. Perhaps Xraytonyb is not for you.
@@peanutbutterjellyjam2179 Appreciate it, keyboard warrior.