damage to bass amp head?

[Dimitree]

Hi everyone
a friend of mine probably damaged its bass amp head.
It is an Ashdown MAG 300 EVO II.

http://medias.audiofanzine.com/images/normal/ashdown-mag-300h-evo-ii-head-338885.jpg

he told me that he connected one of the two rear "speaker outputs" to a mixer channel (keeping the other speaker output connected to the 4x10" cab), instead of using the "line output" on the front panel.
It worked, he indeed could hear its bass through the cabinet and also on the headphones connected to the same mixer.
But then, the next time, as soon as he switched on the amp (still connected to the cab + the mixer through the wrong output), the amp started making what he describes like a really loud and high pitched whine.
The master volume knob doesn't make any effect on it.
The bass signal can't really be heard below the whine, since the whine is huge in terms of volume.
Removing the connection between amp and mixer doesn't make any difference.

I'm not phisically close to my friend so I can't check the amp with my hands right now, but do you have an idea about what could have been damaged?

[Dimitree]

update:

using the "tuner/line output", and removing any connection on the rear to the cab, is fine when routing to a mixer, indeed he can hear his bass as normal, without a cab.
So that tell us that the preamp is fine, but not the power amp.
I found the schematic of the power amp:
https://www.docdroid.net/yhsnwv9/ashdown-mag-elec-blue-apc041schem.pdf.html

he will bring the amp to me in the next days. What should I start testing?

[jimilee]

First be very careful, if you've never worked on an amp, you could kill your self, there's a lot of power stored in the amp after you shut it off. I would first check it with another speaker, you can fry some  voice coils pretty easily. The amp says 4ohms minimum so it may be just that. Then trace the audio signal and look for anything that looks or smells burned, caps, etc....again, if you've never worked on an amp, be very careful, or save yourself some heartache and take it to an amp repair shop.


Sent from my iPhone using Tapatalk

[Dimitree]

don't worry, I'm aware of the high voltages and how to handle them.
I'm just not expert about solid state power amps.
I have the amp here with me now, at a first (and basic) check everything looks fine:
- no smell/burned parts or traces.
- power supply looks fine, I got +65V on V+ and -65V on V- (schematic linked 2 posts above)

but now I don't know how to procede.
What should I check in your opinion?

[GibsonGM]

Have you established that the power amp is in fact receiving signal?  At the connector, "PL 1".   I'd use an audio probe there, myself.   Just to be sure.

Then, I might try to follow the signal.  I would be sure my audio probe has a high voltage cap in it!   

This is the kind of situation where I WOULD be ok with 'easter egging', probing around a bit, and if I found a transistor acting oddly, would check the caps around it.    This "whine"...sounds like an oscillation, right?   So my gut says "a cap somewhere is blown"...perhaps in the power supply...the voltage may be there, but the filtering may be compromised....


Since I personally can't say what would have died under the conditions he told you about (plugging into the mixer), I would not be above testing the transistors' diode Vf characteristics.   It may point out an open.   

Just a shot in the dark.  Not a good way to go about it if you were a tech working for pay, but I'd poke thru for a friend (and have, LOL).

[PRR]

Basics on a sick transistor amplifier-

What is the DC at the speaker output? If not <0.1V, something is wrong. Suspect blown output transistors.

I would note that 150 Watts per pair is very bold for a tranny amp. And there is NO output device protection, very odd for 2003. So fried output devices is likely. BUT when they fry the drivers TR13 TR8 try to drive the load, and melt. Often the Vas stage TR7 joins the rush, though here TR7 has protection for momentary (not sustained) shorts. So if it isn't healthy, it often has -many- blown transistors.

[GibsonGM]

I ran into the same with an old Peavey PA amp I kept running for a friend.  A CS800, could never find out WHY it kept eating the output T's....it just turned into 'I'll change all the power transistors and out the door you go' thing, which kept him happy.  It was cheap enough to do.

He would go play his Olde Home Days thing, break it again, and bring it back next year...Sounds like you may have some trannies to replace...

[Rob Strand]

A CS800, could never find out WHY it kept eating the output T's.

Probably the TRIAC speaker protector.    I've seen that ckt fail in a commercial hifi amp
even though the amp itself was OK. Saves the speakers but it is quite an aggressive thing to do.

[PRR]

> Probably the TRIAC speaker protector.

I do not see that on Dimitree's Ashdown.

It does have a very strange relay+cap power-loss thing which I assume is a mute. If all DC voltages are good but audio won't pass, lifting one end of C7 is a thing to try (if that is a "fix", relay or FET+BJT doodad has gone bad).

[Rob Strand]

I do not see that on Dimitree's Ashdown.

I was only commenting on the CS800.  It's a weird setup.

The voltage measurement you suggested previously is a good place to start.

It does have a very strange relay+cap power-loss thing which I assume is a mute. If all DC voltages are good but audio won't pass, lifting one end of C7 is a thing to try (if that is a "fix", relay or FET+BJT doodad has gone bad). 

But yes, I'm sure that's a mute.  Looks weird to me too - a whole relay set-up to do a mute.   (If you are going to put a relay in why not disconnect the speakers and have a speaker protection ckt like HIFI amps do?  What they have can use a cheaper relay ... shrug)

What you suggested here is good too.

Maybe the speaker impedance should be checked?

[Edit]
At this point I can't see doing what they did causing the amp to blow-up so best to see the voltages.  Also I don't know exactly what that amp does with the speaker out jack.   It's not clear from the schematic as there's no jacks or speaker switching.   If you connect the speaker outputs back to an input there's a chance it oscillated it head off and if the speakers don't disconnect the full-level oscillation would have fried the speaker in no time.
http://medias.audiofanzine.com/images/normal/ashdown-mag-300h-evo-ii-head-338885.jpg

OK looks pretty darn simple.   The outputs just connect in parallel and out the back.

I can't see how connecting to the mixer would affect it unless a significant DC current was flowing (due to a transformer input on the mixed) on top of the AC.  So everything seems like it was working but the o/p transistors were copping a beating.

Maybe the fault is totally unrelated to plugging it into the mixer.

[GibsonGM]

I was only commenting on the CS800.  It's a weird setup.

Yes, it is!  I also replaced the temp. sensor, to no avail, it still 'goes wonky'. Next time I'll look at the triac, thanks....altho the thing's a walking pile of nicotine and maple leaf junk now, ha ha...I will stay on topic now, this is wandering...

[duck_arse]

does SW1 mount on the heatsink or the power Tx, or somewhere else? any voltage on RL1-B?

[Dimitree]

Thanks for the help guys.
Just to male sure before I begin those tests.. should I have a load connected to the output while testing or not?

[R.G.]

No - don't connect a load to the outputs. Solid state amps that work correctly do not die if there is no load connected. And solid state amps that do not work correctly don't get worse because there is no load there. That's a tube amp issue.

In testing large DC coupled solid state amps, the first trick is to make it not eat transistors and fuses as soon as you turn it on. It helps if you have a light bulb limiter to stop things from self-disassembling as soon as power is on.

If you don't have or can't make a light bulb limiter quickly, then:
1. Ohmmeter test the output transistors for isolation from the heat sink - short to heat sink = death
2. Ohmmeter test the output transistors in place for dead shorts from collector base and collector to emitter.
3. Likewise, test the drivers.
4. If no dead shorts found, remove or disconnect the four output transistors, and temporarily tack solder a wire from pin 1 to pin 3 on TR16's empty position, and TR11's empty position. This lets the drivers (TR13 and TR8) act as a low power output stage, and the amplifier ought to work correctly into no load if the bias does not kill the drivers. To prevent that, temporarily short pins 1 and 3 on TR15. This forces the Class AB bias to zero, and makes the amp run Class BC. It will have some crossover distortion, but that's fine for the moment. Now you can check the amp for whether it puts out zero volts DC and passes audio signal without the output transistors detonating. You'll probably also want to disconnect that silly muting relay, or disable it by removing TR9 or R24.

With this set of things done, the amp should still come up and produce an output that is within 100mV of 0V at the "output", and pass audio, perhaps with some crossover distortion. If it does not, you have to start tracking back through the input and predriver stages to find out why.

5. The amp is a strange mixture of clever and "why did they do that?" sections. The inputs are darlington'ed, which is odd. Generally you can get enough gain in a single device to do this well. Perhaps it was because of the 65+ volts the inputs had to withstand from the power supply. The "current source" on the input bases is the simplistic zener and resistor type, when a tap off the constant current for the pre-drivers would have done a much better job. Modern amps would use a current mirror in the collectors of TR1 and TR5 instead of the unbalanced resistor R8.
It seems like using a high gain transistor pair for TR4 and TR2, and making them the inputs, while moving TR1 and TR5 to be cascode devices to hold off the high voltage would have been better, but that's just me.
Then the voltage amp/predriver TR7 is current limited by TR6 (good!) and TR12 is current-sourced. This is good practice.
TR15 as an active bias device is normal practice, but the connection of R21 and R25 to the emitter of TR16 is odd. It looks like it's partially trying to servo the voltage across R20 and the BEs of TR8 and TR11. I'll have to think about how that works.

[Vitrolin]

depending on how your friend connected the amp out put t his mixer,and the volume he used he might also have a problem with a channel on his mixer....
200W@8ohm=56V_p
100W@8ohm=40V_p
  50W@8ohm=28V_p

[Dimitree]

measured the voltages on each node of the schematic,
this is the results:



where I wrote "1V / 0.5V" I mean that it is not stable but oscillating

between this range.
btw I noticed that they float a lot over time.. after some minutes,

all those -0.65V became -0.2V, then +0.04V, and so on

does SW1 mount on the heatsink or the power Tx, or somewhere else? any voltage on RL1-B?

yes, it is mounted on one of the two big heatsink

1. Ohmmeter test the output transistors for isolation from the heat sink - short to heat sink = death

all four output transistor have the central pin shorted to the heat sink.. I measured it with the DMM in continuity test, and it bleeps! but when the unit was not powered. Should that still means something?

[Eb7+9]

assuming your voltages are right the Vbe multiplier appears to be dead ...
start by replacing TR15

[anotherjim]

I measured it with the DMM in continuity test, and it bleeps!

When that happens, the next step is to switch to a low resistance measurement range. Is it a short or a low resistance that should be there? DMM's vary on how close to a perfect short the beep sounds. Could be under 200ohm.

but when the unit was not powered.

Correct. Power MUST be off when testing continuity or resistance. Also, you MUST not have any residual charge left in any capacitors. It's a good idea to voltage test first to check for 0v between the 2 points you intend to test. Any voltage in the circuit will either add or subtract to the meters' own test voltage that it provides to make a continuity or resistance test.

and it bleeps!

A continuous sound or momentary? Circuit capacitance can be enough to produce a short bleep which does not necessarily indicate a fault. On the contrary, it can be an indication that capacitors in the circuit are functional.
Continuous beep definitely indicates a low resistance, but not necessarily 0 ohms.

[Dimitree]

Resistance between the collectors of the 4 power transistors and their heatsink is less than 1 ohm, so I'd say short. When not powered, with no residual voltages.

I think I'm going to take a brute force approach and replace every single transistor on the board. Do you think I need to change some capacitors too or something else?

[anotherjim]

I'm puzzling over how you could have +/65V on the collectors AND have them short to ground! I'd take those power transistors out one at a time and diode test the junctions all ways.
Are they in the package with plastic all over the top like this?