Troubleshooting in a home-brewn amplifier

[Odysseas]

Hello everyone,

I have built myself an audio amplifier, based on a project from Randy Slone's 'High-Power Audio Amplifier Construction Manual', added some stuff on my own (a preamp and a headphone-amp, both incorporating an OPA2134 each), but the amp is a bit troublesome. One toroidal transformer with 500VA powers the big amp (2x10.000 uF per rail), a small transformer with 4.8VA powers the preamp and headphone amp. The big amp uses a 2SK1058 and a 2SJ162 per channel.

The headphone-amp gives me decent performance, but the 'big' amp distorts heavily after bass-hits. For about half a second, there just coming loud, ugly noise out of the speakers, more precise, quite deep, low frequency noise. Afterwards the sound seems fine, but the problems re-appeaqrs with almost every bass-hit. Any idea what that might be? It has been suggested, that I am feeding the amp with more input-voltage than it can handle.

There is another problem, which is probably quite trivial: I have some popping in the speakers, sometimes when I turn the amp on, and sometimes when I turn it off. Once there was an electric spark visible, when the relay for the speakers opened, and the speakers popped. Can I treat this with a few ceramic-capacitors across somewhere?

Here are some pictures of the amp, 4 and 5 will probably be the most useful: **broken link removed**

Thanks for any help!!

 

[Nigel Goodwin]

I'm not familiar with that amplifier, but it's been horribly assembled! - what are all those wires?, and why aren't they nice and neat?.

The whole mess looks like it's going to be unstable, try a scope on it, see if it's oscillating.

 

[Odysseas]

That's my wiring technique It used to look far more terrible, before I re-arranged it all...

Should I apply a signal on it when testing for oscillations?

 

[Nigel Goodwin]

That's my wiring technique It used to look far more terrible, before I re-arranged it all...

Should I apply a signal on it when testing for oscillations?

Try it without first, then try adding a low level signal.

 

[Paul Obrien]

If it is only the bass, maybe the power supply should be looked at. The surge in power at bass freq is obviously affecting the amp, so make sure the BIG capacitor are properly installed. I also noted that you are using a torodial transfromer and these are very good but if there is a cureent in-rush limiting circuit this could also affect preformance.

 

[Odysseas]

There is something before the toroidal transformer, it is a readily available circuit that is supposed to kill the current spike when applying voltage to the transformer. But it is actually supposed to stay quiet after a few milliseconds.

However, this problem used not to be there before I installed the preamp, I forgot to mention.

 

[JimB]

Es tut mir leid, aber...

That is the worst wiring I have seen for a very long time.
There is no obvious separation between circuits, you have unscreened wire on the pre-amp inputs which is very near to the speaker outputs.

You mention a circuit to "kill the current spike...." usually that is a resistor in series with the transformer. After a few seconds a relay contact shorts the resistor.
If the relay contact is not closing correctly, that could cause your problems with bass sounds, the voltage regulation will be very poor.

The heatsinks are doing nothing, there is no room for a flow of air across the cooling fins.

The fans, do they both blow, or both suck. Does one blow and one suck?
Whichever way, you will not get a proper flow of cooling air through that cabinet.

JimB

 

[audioguru]

The problem occurs only if you connect the preamp? Then don't blame the power amp.

Attach the schematic of the preamp for us to see what is wrong with it.

 

[akg]

along with above comments ... is your preamp magneticaly shielded ? . and place it at maximum distance from the power transformers with minimum signal cable lengh. also shield all signal lines . checked for proper earthing ?

 

[tunedwolf]

WoW is all I will say about that build...

The power amp heatsinks should have the fans...otherwise it's pointless...
The wiring is lying in stray fields all over the place, and it's not shielded where it needs to be...
You go to all the trouble to glue up your cable ties etc...but park everything else in there like a spagetti vomit on a saturday night...
Personally I'm not surprised you are having trouble with it.
Does it not mention anything in the manual you read about construction and wiring etc ? If it does, I can't believe they wanted it to be done this way...

I admire your enthusiasm with regard to building your own amp, but in order to be really proud of what you have achieved, you need to pay the same attention to every detail of the build from the metalwork, to the wiring and everything in between.

Sort out your entire build, then diagnose any faults you may have with it's construction or design.

Whenever I build equipment, no matter what it is, I start off by thinking if this bit goes wrong, or that bit goes wrong, how easily can it be serviced. Once you have chosen your easiest route to servicing everything in the box, that will determine where each bit should go, only then can you make the odd compromise based on other factors, like the preamp and the power supply are right next to each other, or this bit or that bit needs screened etc.

as an example of what I might have done with this particular build...

put the power supply right in the middle, at the back of the case with large meshed vent holes in front of it. and put a u shaped shield round it with a hole in the middle at the front for the wiring to pass through...arrange that the shield does not go all the way to the back, but rather leaves an air gap.
The power amp modules I would have put on either side of it with the heatsinks towards the back of the case, so that they are at least level with the power supply shielding, the fans would go in each corner at the back, and be blowing outwards, thereby drawing cool fresh air in through the vents across the heatsinks and back out through the fans...everything is kept nice and cool then. Another shield the full width and height of the case gets put right across the front of it all with the pre amp in front of that...etc etc
this gives you somewhere handy to hang all your wiring from...
All of the signal wiring would be shielded, all of the heavy power supply wiring would be routed as far away from the preamp as possible, both would take a different physical route to their destination and anywhere they cross each other should be at right angles. Speaker outputs would be at the rear, the returns for which would be as short as possible from the common ground point at the power supply as would all other ground points etc etc

I could go on and on, but you get the idea...

Now go away and have a rethink on your design...

I know I'm being very critical of your build, it's true, but don't take it personally, I'm actually trying to help you understand that there are better ways to go about this sort of thing...

 

[Paul Obrien]

Is that the pre-amp power supply with the 24V relay? I noticed it has its own 240V supply, are you sure you haven't disturbed the inrush supression circuit.

 

[audioguru]

It looks like a steel bolt is through the center of the torroid transformer. Doesn't it make a "shorted turn"?

The transformer is 500VA but I didn't see a 300W amplifier there. I saw a couple of 60W amps.

 

[tunedwolf]

it would only be a shorted turn if he were to have a wire completing the circuit to what the transformer is bolted to...
but I do agree it looks a bit on the big side for what he's powering...

 

[Odysseas]

Okay, so many responses, I try to isolate the problem systematically.

The pictures are not perfectly up to date, the signal wires are now double shielded, and the signal-ground is now connected directly the preamp-board (middle, just below the big pot). I think we can eliminate that. However, the preamp-board is not magnetically shielded, do I need to do that?

As far as I can tell, earthing is good. The HQG-point is located quite in the centre of the case, and I measure low resistance all across the place.

I agree the transformer is quite big, I will probably remove it some time in the future and use it for another project. But right now, I want to get the amplifier to work properly. Or could the big transformer cause any problems?

Here is a schematic of the preamp (mono, multiply it by four and you will get what I have): **broken link removed**.

Yeah, there is a steel bolt in the middle of toroid and a disc on top of it, but it is not a closed circuit.

The fans both suck, and the air is 'forced' through the heatsinks, I drilled a number of holes in either sides of the case.

@tunedwolf: You enjoy talking other peoples work bad, do you? Of course, you do. I know those folks like you, you get a big laugh out of it, one could even say that you derive pleasure from it - I know that exactly. Well, I guess there is nothing I can do against that, but your speech has failed to make me lose my belief in my project. It also has not managed to intimidate me. In any case, you can expect me to ignore you in the future. Please don't take this personally

 

[audioguru]

Your preamp is just an adjustable gain stage. It "causes" the problem because it allows the amplifiers to produce more power when its volume control is turned up.

As others have said, your power supply voltages must be collapsing during high power bass notes. Probably caused by the circuit that reduces the turn-on surge.

 

[Odysseas]

Well, I will see if I find sufficient space in the enclosure, then I will try to built a small circuit that kills the power spike myself. However, I have some difficulty in thinking that this is a cheap part that only causes problems.

@audioguru: I am afraid I do not quite understand your explanation. Does this come down to the point that the preamp is apparently 'clipping' the bass-peaks off?

 

[audioguru]

You said the problem only happens when the preamp is there. I think the problem happens only with the preamp because it can turn up the volume to where the power supply voltages collapse during the louder (higher power) bass notes.
Without the preamp then the power is low enough so that the power supply's reduced voltages aren't noticeable.

Simply connect a piece of wire across the current spike reducer and see if the distortion problem disappears.

 

[tunedwolf]

hey Odysseas, I wasn't picking on you, personally or otherwise.
Sure I said your amp was a dogs breakfast, but that's because it is, I put that down to your inexperience with building such a project.
I don't derive pleasure out of putting anyone down, in fact I was actually complementing you on your enthusiasm for even tackling such a project, as many wouldn't. I was hopefully showing you a better way of building such projects, based on my own experience, so that the next time, and I do hope there will be a next time, you build an amp you will benefit from what you have learned this time round. If you choose not to make use or benefit from the wealth of experience, freely shared I may add, from the folk here, that's entirely up to you. I will be just as happy not comment on anything else you post in the future. I am sorry if I have caused you offense in any way, it was not my intention.

 

[Nigel Goodwin]

hey Odysseas, I wasn't picking on you, personally or otherwise.
Sure I said your amp was a dogs breakfast, but that's because it is, I put that down to your inexperience with building such a project.

You were perhaps a little caustic?, but everything you said was true!.

Before trying to fault find the amplifier really needs rebuilding properly - it's not just a cosmetic thing, it's built so badly it could hardly be expected to work?.

 

[tunedwolf]

Someone has to point him towards the light...before he tackles something possibly life threatening with the same attention to detail, I guess I appointed myself to the role...it's an old habit I'm trying to get out of...
No matter, no point in me worrying about it, it's his amp, his workmanship and his pride.