Welcome to Solid State Guitar Amp Forum | DIY Guitar Amplifiers. Please login or sign up.

March 28, 2024, 01:39:33 PM

Login with username, password and session length

Recent Posts

 

Bogen GS-150 Schematic, and Help Needed

Started by paelgin, December 31, 2014, 08:24:46 PM

Previous topic - Next topic

paelgin

I bought a broken Bogen GS-150.  No output from the power amp, although the pre-amp seems fine (seemingly proper signal on pre-amp output).

In the power amp section, I have +/- 43 volts (closer to 50) rails and no obvious shorts or burns. In diode checking with a Fluke diode voltage-drop tester, almost nothing jumps out.

Except D300. It shows 0 volts drop across either direction (in-circuit testing only), and has only about .2 ohms of DC resistance, which takes the power amp output straight to ground.

What is D300?  I do not have any other manual, and cannot tell what D300 is. In reviewing the schematic, I thought that it looked like a back-to-back zehner to protect the transformer from over-voltage on the output of the power amp. I'm tempted to remove it to see if I can get output (since I have not broken out my scope).

Any thoughts would be appreciated.

Phil in beautiful Talking Rock, Georgia, USA.

paelgin

In a moment of clarity when re-reviewing the schematic, I realized that the output from the power amp goes to the transformer (which is connected to ground) through a "link" added on the terminal strip which is parallel to D300.

Since the transformer can be taken out by removing the "link", I took the link off then re-tested D300 (dry, in-circuit). Now it looks pretty normal to me with a climbing voltage drop as the caps charge in the associated circuitry. The DC resistance across D300 is a little over one meg-ohm.  Seems fine to me. 

Back to the Fluke meter and amp.

J M Fahey

#2
Yes, D300 is most probably a couple back to back Zeners in series.
Maybe 50V or so, enough to stand normal audio output peaks but not beyond.
Think will have , say, 5V extra peak voltage beyond rail to rail.

Looking at that schematic, the amp has many points where signal muting is expected, under certain conditions, even a remote Master volume conttol, a limiter, lots of switching, so only safe way is to inject signal at some input and trace end to end to see where it stops.
And then we find why.

paelgin

Thanks for your reply. 

My audio tester is in some box in my garage from my last move. I'll have to see if I can find it.  Once I opened the amp up and found nothing burned, I kinda' assumed that's what I would need to do. 

But since I had audio going into the power amp, I thought I could just focus on the power amp, and I didn't see too much signal manipulation in those stages. So I started my dry testing (no power) on the semi-conductors, to no avail thus far.

If I make progress, I'll report back.  I do not know the history of these (I have 2) and since I found several discrepancies between the schematic and actual board, I may have a revised board, or revised schematic, or maybe someone else did some "repairs" (although the board looks virgin).

Phil in beautiful Talking Rock, Georgia, USA (until I move to gorgeous Young Harris, Georgia, USA)

Roly

Gad!  There's a lot of it, isn't there?

I'll just note that opening SW8 "In/Out" gives you direct access to the power amp input.

It may have more cylinders than most, but it's still same old same old.

Power it up, no load, internal OPT UN-linked, via the mandatory Limiting Lamp.

Check that the supplies are roughly equal, and that the half-rail is within a few hundred millivolts of ground.

No - diagnose why.

Yes - load tests, light low drive first, moving carefully up.


{Ah yes, the mysteries of variations during production.   :loco }
If you say theory and practice don't agree you haven't applied enough theory.

J M Fahey

Something different from standard SS amps :
1) this one needs a load connected, because it has an output transformer.
2) no signal DC offset voltage needs to be 0 to +/-2mV , because, you got it ... it has an output transformer.
It says so in the schematic and provides a trimmer pot to adjust that.

That said, it has the provision of disconnecting it :)
Open the top link, the one living between "direct 4 ohms" and "70V" out and connect your speaker or load between "direct 4 ohms" and "Common/gnd"

3) lots of signal killing points available; which does not mean there's something wrong, simply one or more may be activated and you don't notice it, that's why a scope is your friend.

If you don't have one, work backwards, step by step.
Start injecting signal straight in the power amp and listen with your speaker, then proceed towards the input.

Some muting/signal killing points (I may miss some though):

a) Sw8 "In/Out" .
That said, start by setting it Out and inject signal in "Power Amp in" which lies after it.
Hear anything?

b) R164 Master Volume set too low or 0 .

c) Remote control IC14 enabled (it's an optocoupler) shorting Master Volume.

d) Sw7 may be off, grounding "Aphex" processed signal.
This will not kill all sound, but ssssssssssssibilansssssssssssssssse will dissapear.

e) SW6a/b enables/disables different combinations of (graphic?) Equalizer and Aphex processing.

Some of these need main signal killed so the EQ/processed/both audio can be heard.
Now if you select one of these modes but then do not use it and set its control(s)  to 0, then you are left without main or processed signal = silence.

f) AFAIK Q11 is some kind of mute, controlled by SW10 .
Edit: it's a partial mute, it's a low cut filter.

Ok, got tired and didn't even reach half the possible mutes which might be enabled.

This is a wonderful amplifier, incredibly versatile with tons of options, if there was just one amp left on Earth it should be this one, but that makes it also very complex.

g)

paelgin

Thanks again for the suggestions. The +/-43 volt rails are a great place for me to get back to.  When I 1st fired one up, I focused on the possibility that the previous user had incorrectly set the switches/mutes/link or connections. 

With a stereo input on Aux1/Aux2, I was getting signal out (varying by the Master Volume setting) at the Pre-amp output RCA connector, but nothing on the 8 ohm output. When I switch SW8 "In", I also got the same output on the Power Amp In RCA connector, so the pre-amp (with associated mutes and seems to be passing signal.  That's why I was concentrating on the Power Amp section, and was surprised to find that none of the transistors, diodes or passive devices show any sign of heat. I have to assume that the signal path is being diverted to ground or something early in the power amp, otherwise there's lots of power that has to be dissipated, but nothing looks burnt.

Phil in beautiful Talking Rock, Georgia, USA.

Roly

Quote from: J M Fahey1) this one needs a load connected, because it has an output transformer.

But if you...

Quote from: RolyPower it up, no load, internal OPT UN-linked, via the mandatory Limiting Lamp.

...you can service it just like any other OTL power amp - the OPT isn't in the NFB loop or anything.

Quote from: paelginotherwise there's lots of power that has to be dissipated, but nothing looks burnt.

Quite so, so we can conclude from that that the power stage isn't being driven and shorted or there would be all sorts of signs of distress, so the input signal must either be shorted or open and thus not getting into the output stage.

Past time to fire up the 'scope for a bit of signal tracing.  :dbtu:

My attention is drawn to the SW8 area because you appear to have good preamp signal up to about here, Pre Out, yet there seems to be no signal finding its way into the power amp input.

As the output half-rail appears to be close to zero volts it is reasonable to assume that the power amp itself is actually okay.

Maybe I missed it, but have you fed a signal in Power Amp In and got good output?
If you say theory and practice don't agree you haven't applied enough theory.

paelgin

Thanks for your thoughts.  I will have to get some equipment out (finding it may be hard) to do more checks, although I will (after re-assembly) fire up the amp (with a speaker on the 4 ohm and ground terminal strips and OPT link removed) and verify rails. 

I had it running for probably 15 minutes to no avail when I first got it as I made sure all controls, switches and options were set to allow output. And no magic smoke escaped.

When I put input on the power amp in, there was no output.  With the stereo input on aux 1 and aux 2, the preamp out had a seemingly line level output that varied with the volume control.  To verify SW8 (Pre-amp to Power-amp enable), I verified that the PowerAmp input RCA jack also had a signal (seemingly but unmeasured line level).

In the schematic, something has been bothering me.  D9 and D10, (which look to me in the schematic like voltage dropping diodes for biasing the bases of Q35 and Q36) are installed backwards from the schematic.  On the board, the anode of D9 and the cathode of D10 are connected to ground. 

It occurred to me that if they had previously failed, maybe someone replaced them (and maybe other stuff), but I can't tell from the solder: maybe it was a professional repair, not the sloppy job I usually do with my 30 year old 150 watt Sears soldering gun. But what if they were/are installed backwards? Kinda' looks to me as though they are part of some sort of feedback circuit, although I have been out of electronics for so long and only understand basic stuff. But it looks to me as though Q38/Q39 form Darlington pairs with their respective output transistors to achieve the current gain, and the base of Q35/Q36 is driven partially by the emitters of those Darlington pairs. (Hopefully, I won't have to learn push-pull class D or whatever amplifier theory).  When I first read the voltage drop to ground and it was backwards from the schematic, I chalked it up to updated/corrected schematic that I did not have.

Well, back to real life.  I don't know if I'll have time to do anything else on this before our trip to the CanOpener Airstream Rally in Santa Rosa Beach next weekend.

Phil in rainy, stormy Talking Rock, Georgia, USA.

phatt

Often missed but my guess is that you may find the preout/powerin relies on a pathetic switching socket and they are a common failure point, Cheap nasty 6.5 sockets now common in even the best equipment. :grr :grr
Hard link the circuit with a short cable and see if it makes music before you dive into the dark secrets of a rather complex circuit. :-X

I own a very similar amplifier (different brand name) and it had heaps of niggley little issues.
Gutsy unit but mine runs quite hot.

BTW, D9 D10 are part of the protection system, along with Q35,Q36, D7, D8.
I think you will find Q31 sets the bias, with 2 resistors across the base.
Phil.

Roly

Quote from: phattthe dark secrets of a rather complex circuit. :-X

Yeah, there's a lot of stuff in there alright, but we don't want to be intimidated by its expanse; it's just a large collection of small stages, and a bit of step-and-repeat like a powered mixer.

I'm tending away from the typical faulty Fx socket switch because a) it looks from the cct like they are RCA connectors, but b) mainly because paelgin has mentioned a couple of times that with SW8 closed he was getting signal at Power Amp In.


Paelgin - Before you connect a speaker (or the transformer) you need to make sure that the output offset is low, in this case should be within +/-2mV of ground.  This may be hard for you to measure, but if your DMM shows more than a few units of millivolts you should be suspicious.


D9 and D10 are bias diodes for the output stage Safe Operating Area (SOAR) protection circuits around Q35 and Q36.



For this device the top line defines the absolute device current limit, then it slopes down according to the device power limit, next is the limitation caused by "2nd breakdown" risk, a failure mode that only occurs when the voltage across the device is high and the current flowing is also considerable - basically you get runaway current concentration through a small area of the whole chip, which melts through and shorts the device.  This is also called Lateral Current Instability and modern power devices are generally laid out to make this much less likely to happen.  Finally is the maximum voltage limit.  This is the SOAR curve for this device - safe operation of the device is only assured within the shaded area.

A resistive load would produce a straight line Load Line, say between 16 amps and 120 volts, but real world loads, speakers, are electrically reactive components, inductive or capacitive, and at different frequencies will draw current out of phase with the applied voltage.  This makes the load line expand into an ellipse, and this may very well pass outside the shaded area, endangering the devices, so SOAR protection is applied to allow some limited simultaneous high current and high voltage operation.


Quote from: PaelginOn the board, the anode of D9 and the cathode of D10 are connected to ground. 

Ah ha!  But we need to be very careful here that the board is exactly like the circuit, that D9 is indeed in the +ve half of the OP circuit and D10 is in the -ve half, but if it is it looks like these have been replaced backwards, D9 should have the cathode (band end) grounded, D10 etc.  A possible effect of this error would be to turn on the protection transistors Q35 and Q36, effectively shutting down the output stage.  This would be consistent with input, no output, but no smoke (and the offset is okay?).  It's not the "short across the signal path" that I expected, but it looks like it would fit the bill.

Q35/6 act as clamps on the incoming drive to Q38/9.  These sense the current flowing in the output stage by measuring the voltage across a couple of the emitter resistors (R202/3), but because the diodes are connected to ground rather than the half rail the current sensitivity changes with signal amplitude - the output transistors require more protection against high output current and high voltage than for high current at low output voltage. (R193/3, D9)

Start at the main +ve supply and trace down through Q38 to R195 to Q35, to R194 to the anode of D9.  If you come to the cathode (banded) end of D9 then they are in the wrong way around.  (and you might check D7 and D8 too - all bets are off at this stage).  Ghod knows what has been going on in here before you came along, you could find anything, your light bulb limiter is a must just for the moment.

Q1. are the diodes in the wrong way around?  {I think they are, but you need to check very carefully.}

Q2. if/when corrected, is the half rail still near ground?

If so you can try a low level signal test and see if you are now getting signal on the output half-rail.

Again, if so I'd be inclined to call that a fix and a win.


{note that if these diodes are indeed in the wrong way around there is a much higher than normal possibility that something(s) else(s), in fact just about anything(s) else(s,) may be stupid too.  Be alert and suspicious.  I hate amps that have been hacked badly because it is normally impossible to understand WTF the modifier had in mind when they did that - only that it's electronically misconceived.  e.g. you replace both the diodes in the protection circuit (why?  nothing seems fried?), and you get it wrong, it now/still isn't working, so you do what, abandon the amp?  Horribly Hacked amps like this turn up from time to time, and frankly I'd much rather be the first man in.  In my industrial career I often got called after a few others had had a bash, and it was not uncommon to take four days to unscramble their omelette before I could get down to the original fault, one day or less.}


I'm curious how much you paid for this "not going" amp because I have a thesis that these are often a very good risk.
If you say theory and practice don't agree you haven't applied enough theory.

paelgin

Thanks again for your thoughts, ideas and corroboration of some ideas. 

I bought two (non-working, have only looked at one) for $13 plus tax. People (potential bidders) did not want to pay the big bucks to ship, and I could do a local pickup.

When I have time to look again, I will take it apart again and circuit trace the diodes.  When I use my Fluke diode test, I get a .609v drop on each, but definitely backwards from the schematic. I will have to check D7/D8 as well (see picture, all four look kinda' irregular).

Phil in cool Talking Rock, Georgia, USA

Roly

Quote from: paelginI bought two (non-working, have only looked at one) for $13 plus tax.

YO HO!  Can't lose then, and will end up winning big time, two for $13(+tax) and a bit of hair.  I love buying gear "by the pound".  :dbtu:


Rats!  D7 at least seems to match its PCB overprint symbol (the overprint could be wrong, but it's very unlikely).  Bah, would have been too easy.


Hummm.  At this point I would inject a signal to Power Amp In (radio, cassette, MP3 player; it's not critical, just something) then starting at the power amp input R170, I would follow it along the signal chain, Q26, Q30, etc, with the CRO until it stopped, then work out why.
If you say theory and practice don't agree you haven't applied enough theory.

samguitar

I know that this is really late, but if you are still having problems with a Bogen GS150. Your problem is probably the soft start caps. Change C121 and C122. Upgrade to 50V caps.
This fixes 99% of units with this problem. I know I work on them everyday.

paelgin

In moving, I threw the dissassembled one back together, but managed to lose some tiny screws for the faceplate. Now I'm in our new home but have not gotten back to this.

Thanks for your insight. I will see about the caps C121 and C122 when I get to it, but I'm in no hurry.

Phil in gorgeous Young Harris, Georgia, USA