Interesting, I'll look into this... I've never used spice. All components around IC9A are within tolerance so there's nothing unusual going on there. In the next few days when I've time I'll check out IC9B again and up the comparator thresholds to 3V and see what difference that makes... 🤔
Thanks for doing that and it looks really interesting to check out ... I've looked at spice before years ago but I don't know how often I would use it in repairs and also being on Linux doesn't make that any easier.
Using spice looks great from a design point of view but can you give me some hints about how I could use this to find the problem I have here?
Last edited:
I am on Linux as well and use LTSpice IV together with WineHQ - works for me for years.
The simulation just helps to understand the circuitry.
To trace your problem I would check the output of the current sense circuit with a sampling scope, triggering on a max point trying to find the overcurrent glitch. I suspect that this all works fine and that by playing the bass you produce some subsonic pulses that overload the amp. This specially may occur in case your preamp clips/distorts/compresses the signal.
The simulation just helps to understand the circuitry.
To trace your problem I would check the output of the current sense circuit with a sampling scope, triggering on a max point trying to find the overcurrent glitch. I suspect that this all works fine and that by playing the bass you produce some subsonic pulses that overload the amp. This specially may occur in case your preamp clips/distorts/compresses the signal.
That's what I've been doing .... this is a high quality commercial production bass guitar amplifier combo speaker/cabinet that worked perfectly for many years ...
The current sense circuit has been scoped & components checked out over and over right down to replacing all the ICS. Also there is nothing out of place with the signal before it reaches the input of the 40pin 600W amp module detailed on the schematic but the mosfets on that seem fine and the gate drive waveforms look normal. Whatever is causing this happens in the amp module circuit or after that but I have been back to this several times over the last few months and cannot pinpoint the reason. I am just a repair man though and have never designed Class D amplifier or current protection circuitry which is why I am here seeking help.
Telling me to analyse the circuit is just going back over what I have done already ... if you think I'm doing it wrong please detail your steps and I'll be happy to follow them if you think it will provide the solution ... 🙂
Just to save time I should add that I have been in contact with the guy that designed the pre-amp for many years now for other reasons and he has provided me with all the test and service info to fully check out the pre-amp as at first I suspected it was some kind of clipped harmonics exciting the amp filter resonant frequency because initially it only happened on D notes .... I have scoped and tested the pre-amp following his instructions and it all checked out so focus then concentrated on the upstream circuitry...
Last edited:
It's a TE715x (like the photo above shows on the front plate) 🙂 ... but there are a lot of overlap of designs and many changes to each model so there is a version of the 1215 with this same powersupply+amp module unit but what I have here is a 715x.
Could you explain in detail your suggested modification please?, I would have thought that doubling the gain into IC15 LM393 (which I thought was a kind of window comparator) would have made the tripping worse? Am I wrong about this?
I really want to understand how this circuit works from the sense resistors after the LC filter on the output of the amp module to the output of IC15 which is triggering the power supply shutdown, I've had a lot of people saying they think it is working properly ... but nobody yet has explained how it works which is what I am looking for?
The objective is to understand what has caused this over current fault to happen.... 🙂
I drew the simulation of that circuit to make it understandable.
You should know basic concepts like differential amplifiers, and read the schematics.
My proposal reduces differential gain thus increases the tripping point.
Otherwise I do not have a clue what the cause of the problem is,
and the developer obviously doesnt either
You should know basic concepts like differential amplifiers, and read the schematics.
My proposal reduces differential gain thus increases the tripping point.
Otherwise I do not have a clue what the cause of the problem is,
and the developer obviously doesnt either
LOL you had me there, I just skimmed your answer saw the "2x330" and 660 was what popped into my head not 165.
Your idea is good and better than changing the comparator thresholds.
The pre-amp designer is not the same guy who designed the powersupply+amp module unit, that was designed by a completely different company.
My attention is just taken up at the moment with some other stuff but I need to get this amp back out again a.s.a.p. ... it's been months since I last looked at it & there are other things I want to check out on it like the input buffer IC1 (MC33201).
I can see now how spice could help in situations like this, I'm gonna try and install it tomorrow. I use AVLinux MX edition which is a debian derivative so hopefully it should be quite straight forward but not got a lot of spare time at the moment so hoping not too much of a learning curve or I'll have to put playing with it off for a while.
Your idea is good and better than changing the comparator thresholds.
The pre-amp designer is not the same guy who designed the powersupply+amp module unit, that was designed by a completely different company.
My attention is just taken up at the moment with some other stuff but I need to get this amp back out again a.s.a.p. ... it's been months since I last looked at it & there are other things I want to check out on it like the input buffer IC1 (MC33201).
I can see now how spice could help in situations like this, I'm gonna try and install it tomorrow. I use AVLinux MX edition which is a debian derivative so hopefully it should be quite straight forward but not got a lot of spare time at the moment so hoping not too much of a learning curve or I'll have to put playing with it off for a while.
I should mention that in my simulation there are two single 10mR shunt resistors, but in the schematics you will find 2x2x10mR shunts. resulting in 5mR shunts. Based on this my calculated gain was faultiy.
Ok have installed LTspice64 with wine and loaded your .asc but I'm not sure if spice is working properly yet ... I'm gonna have to watch a couple of spice tutorials I think ...
Yes I noticed the single 10mR .... once I understand spice better I will correct.
Thanks again for drawing that .asc out ... hopefully mid or end of week I'll get time to get back to this and I'll report back with updated info from the unit ... 🙂
Yes I noticed the single 10mR .... once I understand spice better I will correct.
Thanks again for drawing that .asc out ... hopefully mid or end of week I'll get time to get back to this and I'll report back with updated info from the unit ... 🙂
I will encourage you to play around with LTspice.
You should start with transient analysis and then place your voltage and current probes wherever you want to see how the circuit is working. Btw there is a long LTSpice thread inside DIYaudio.com.
You should start with transient analysis and then place your voltage and current probes wherever you want to see how the circuit is working. Btw there is a long LTSpice thread inside DIYaudio.com.
I eventually got the latest version of spice to work properly in Linux (still not convinced of it's uses in repair work but definitely a handy analysis tool now I can get it to work). Sorry for no replies, I've had bad food poisoning but will be back when better and have time again...
Sorry about the long delay in follow up I had a bad health flare up for three months but almost fully recovered now.
Anyway amp is working satisfactorily now... It would appear this is a common fault with these amps as I have seen three of them with the same problem but this one was the first one where I could replicate the fault on the bench to investigate. So the conclusion after extensive investigation & testing is that the amp has no obvious broken components and the problem would appear to be a combination of several component values drifting slightly with the age of the amp combined with an over sensitive overcurrent protection circuit. In the end the gain of the differential op amp in front of the OC window comparator was reduced by changing the gain setting resistors to 270k... This had the equivalent effect to raising the OC window comparator thresholds from about 2.4V to just under 3V and cured the four second cut out problem.
Thanks to all who posted, all the input helped find the best solution... 🍻🙂
Anyway amp is working satisfactorily now... It would appear this is a common fault with these amps as I have seen three of them with the same problem but this one was the first one where I could replicate the fault on the bench to investigate. So the conclusion after extensive investigation & testing is that the amp has no obvious broken components and the problem would appear to be a combination of several component values drifting slightly with the age of the amp combined with an over sensitive overcurrent protection circuit. In the end the gain of the differential op amp in front of the OC window comparator was reduced by changing the gain setting resistors to 270k... This had the equivalent effect to raising the OC window comparator thresholds from about 2.4V to just under 3V and cured the four second cut out problem.
Thanks to all who posted, all the input helped find the best solution... 🍻🙂
Maybe the amp is fine and its tripping simply by too big a signal ?
Guitars are notorious for outputting large transients.
Buy a bigger amp.
If the signal gets big enough to saturate the output then that can cause problems.
Guitars are notorious for outputting large transients.
Buy a bigger amp.
If the signal gets big enough to saturate the output then that can cause problems.
This combo's pre-amp has an input gain control with "traffic light" peak transient indication, this was set correctly and the pre-amp output is also limited.
The power amp is quite large ... 600W RMS into 4ohm.