ZPA 0.5 control board problem

[Silverhorse]

until now I couldn't find out why the yellow LED is working for 1- 2 seconds when powering up.
But in the end I think this is the normal operation of the amp. I have several other ZPA's in a daily system and when powering them up I have seen never a Yellow LED lighting up. But then I thought, what happens when I remove the 12v continuous power from the terminals and make sure that all the caps of the amp are empty. Well when I power up after doing this the yellow LED is also lighting up for 1-2 seconds. So the time the caps need to charge the yellow LED is working. Once the 12v continues power is connected the yellow led stays always off. I think I have one problem less...
Can other ZPA owners confirm this?

[Jacampb2]

I checked my 0.3 that I use as a reference amp on the bench the day after I replied last time. This amp is on a power supply and it's completely without power for 12+ hours before I fire up the shop in the mornings. I wouldn't swear to the caps being completely discharged, but I expect they are. I'll double check tomorrow.

My 0.3 does not light the yellow thermal led at any time as it's powering up. In the case of my amp, the remote is switched via a head unit still, so it will not be going high at the exact same time as the power supply comes up. If you have the remote hard wired to b+, that may be why you are getting odd behavior. I'd try connecting the constant power connections and then remote to see how it behaves.

Good luck,
Jason

[Jacampb2]

I can confirm that if A)remote goes high before the B+ or B) Remote and B+ go high together, that the thermal LED does indeed light on power-up. This is probably not a great way to power the amp, and the fact that it does this means that parts of the protection circuitry are having their power supplies come up in the wrong order.

[Silverhorse]

Thank you Jason for checking this out.

did find some time to continue this project.
I pulled out all the output transistors and when I power up the amp the red overload LED is still on. The intensity of the overload LED is going together with the power supply voltage. I use a 10.5v voltage supply so the overload LED is almost out. When you put 13.8v on the supply the LED is full on.

When I measure the voltage between the collector and speaker ground of alle the 4 banks of 6 output transistors there is a difference in voltage.
in the right channel (2 banks of 6 BJT's) I have everywhere the same voltage of 50V (testing power supply = 10.5v) . In the left channel there is a difference. The BJTs with numer Q136 until Q141 (6 time NPN + 6 time PNP) measure 28V the other ones Q236 until Q241 measure 50V. So what is going one here?

[Jacampb2]

Were the voltages taken with or without the outputs in circuit?

When you say speaker ground, do you mean the non inverting output for each channel, i.e. the terminal with no signal present that is connected to secondary/audio ground? I ask because the only one of the terminals marked (-) are actually connected to secondary ground. On the inverted channel the (+) terminal is at secondary ground potential.

[Silverhorse]

the voltages were taken without the outputs in the circuit.

I measured the 50v between the collector and the + and - of the left channel, so the + of the left channel is the non inverting output and the - (min) the secondary ground potential if I understand correctly. In other words, on both the + and - of the left channel I measured 50Vdc + or - depending witch bank.

The problem of the difference in voltage between the different banks is solved. One of the 4 IRFZ44 fets I replaced had a bad via on the collector (Q19). Normaly I always to the beep test on traces / via's after replacing components, but this one I skiped . So now I have everywhere the same voltage. But the overload LED is still on.

[Silverhorse]

tested the output BJT's and al the gate resistors. Nothing wrong with them, put them back in and still the same problem...

[Silverhorse]

today I tested the amp with my test speakers and head unit with the risk of damaging my speakers.

the amp is playing wonderful and powerful without any distortion in the speakers but the problem with the red overload LED stays the same. It's all the time working.
So the most logical explanation (correct me if I'm wrong) is that there is a problem in the overload circuit like you already told me Jason. I have tested and checked all the via's of the vertical boards, no shorts or conducting problems.

Is there anybody who know's in what part of the amp I have to dig in? Witch parts are responsible for the detection circuit?
There is no doubt that there was a excessive load on the 15v supply, I think that there are other components damaged because of this.
How do you find out?

I can borrow a scope if needed...

[Silverhorse]

on the control board resistor R47 is a 100K resistor, on the schematics of Jason and my other zpa's it is a 10K resistor.
could this be causing the problem? I can't verify it now because I soldered out the board. Could this be the reason that the green LED is not going out when the red LED is on?
It seems that the resistor was there from day one so if it left the factory like this that means that the problem was not detected in normal use?

[Jacampb2]

Sorry, I haven't been on here lately. I'm waiting at the vet for my dog at the moment, but I'll try to look into it when I get back to the shop.

[Silverhorse]

how is the protection circuit reacting when R47 is 100K ?

[Jacampb2]

I'm really sorry, I forgot about this. R47 may be responsible for the fact that the green LED is not lighting. It should not be causing the O/L led to light. The weird thing is that if the o/l led is lit, then the power amp should be in mute. The same output that drives the led should cause the protection to trip. I'm wondering if possibly the transistor is leaking, is the led very dim? I'd change r47 to spec regardless and see what happens. Put a lot profile socket on the board for testing so you don't have to constantly rework the joints.

[Jacampb2]

Interesting story, I have a clients MS2125TA on the bench right now. In addition to a mess of other issues, it to has the O/L LED lit all of the time even though protection isn't tripped and the poweramp is not muted. In this amp it turned out to be a bad LM339. It was leaking just enough to allow the led to light, but not enough to actually trip the protection and shutdown the power supplies. If you haven't checked it out on your board, I'd take a look at it. I am seeing more and more of these amps come in with bad comparators.

[Silverhorse]

there is something wrong with the control board but I don't know what exactly yet, need just more time to invest in this project...

[Silverhorse]

did find the time to continue with this repair.
I switched the control board of a working zpa (so I know that the board is 100% ok) with this one. Well... this zpa is now working perfect !
So now I know for sure that the problem is in the control board. What can cause the overload (red) LED burning while there is no overload fault from the transistors?
I changed the 2N4401 next to the red LED but problem stays the same. The KA339 is also changed so what else could it be?

[Silverhorse]

could it be wrong spes of the KA339?
the one I used: 512-KA339A (Mouser)
512-2N4401TA

[Jacampb2]

The lm339 should be pin compatible across manufacturers. It's possible that you got an out of spec one. You'll need to check voltages on the inputs of the comparator on pin 6 & 7. Make sure pin 6 is at primary ground potential. Other than that, I'd check the parts I circled in the attached photo. I'd also check very closely to make sure there are no open traces on the board and that there are no damaged through holes.

Good luck,
Jason