• TalkBass has been independent since 1998. Add your voice.
    Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
    Join free Log in
    Want zero display ads or expanded classifieds tools? Compare plans.

BA115HPT: test points way off - what could it indicate?

O

okwestern

Jun 1, 2020
63
10
#1
My fathers Ampeg suddenly start humming like crazy after playing it for a while, but it he usually turned if off and on, and it would work for a few hours again. Opened it and cap C52 was cracked in the top. Replaced it and then measured bias voltage. Left bank was not a good match (though both had blue dots), while right was perfect:
R73 (Q8): 2.6mV
R72 (Q5): 12mV
R74 (Q4): 6.5mV
R75 (Q11): 7mV

But I chose to measure test points, and possibly change any other parts before fondling with the biasing.
I have quite a lot of DC on the speaker out (105mV), and some other points also quite off. I've written out which style I put it in as the schematics doesn't say, but measurements vary with style (I guess style 5 is used for testing in the schematics?). I think I also found an error in the schematics (TP1 should be measured on the output of the opamp, not the input, as the input should be almost equal to the signal generator output 170mV output).
I've uploaded an image instead of trying to make all the numbers look good in text here.
Any idea what could be the reason?

forum.png
 
#2
Most DC values are measured with no signal present, so be sure you are differentiating this as this will skew the results.

Your AC values are fine, these are very rough values though TP10 is way off and likely an incorrect measurement OR there is too much DC offset beyond what the servo can correct for. U5B is a servo DC offset correction stage that also serves as the diff amp and handles global feedback.
 
  • Like
Reactions: RSBBass, Rip Van Dan and DJ Bebop
#3
Thanks. If I assume all DC measurements should be without signal, which of them should actually be with signal (if any)?

And for TP10, do you mean U3B (as HPT doesn't have a U5 - perhaps you looked at BA115HP)?
I hooked up to TP10 again and have the same measurements.
How do I troubleshoot this?
 
  • Like
Reactions: DJ Bebop
#4
DC measurements on Ampeg amps are generally square boxes, AC voltages are generally round boxes, but each set of service docs will identify this.

How do you troubleshoot this? How much experience/knowledge do you have in troubleshooting global negative feedback solid state amps, and do you have a suitable scope, signal generator and DMM? NFB amps like this are tricky because for every action is an attempted inverse reaction.

U3B is the servo on the HPT, I was looking at the BA-115 schematic. With TP-12 being good and the output of the servo being bad, there is something unbalanced within the power amp.
 
  • Like
Reactions: Rip Van Dan, DJ Bebop and Hoochie Coochie Man
#5
Good to know square vs. round. What about TP14/15, they are square but 0.4Vp-p with a sawtooth symbol, so they should be AC as the DC on this rail is +/-50V? Or how do I interpret the ones with sawtooth waveform?

Not much experience, but as with everything, everyone have to start somewhere :)
I do hardware development for work, so I have decent equipment (1GHz scope, 6.4GHz signal generator and a Fluke 289).
 
  • Like
Reactions: DJ Bebop
#8
okwestern said:
How will the low reading at TP10 affect the amp (sound, amp life, etc)?
And where do I start troubleshooting this?
Click to expand...
Need to understand how global negative feedback amps are supposed to work first, then look for what will cause an incorrect value at TP10. It could be anything.
 
  • Like
Reactions: Rip Van Dan and DJ Bebop
#9
So not understanding any of this but wanted to inject something here.

It's been said that some people don't wanna help people here. Not give advice to help me with my project. You people discourage people on fixing their own stuff and so on.

This is what it takes folks, IMHO.
@okwestern came asking for help but not only asking but indicating he had the know-how to handle the project at hand with the info needed and exhibiting the understanding to complete the work.

This to my mind makes it easier to dispense knowledge in a safe responsible way.
Not in a way that can cause bodily harm at most, or blowing stuff at least.

If one wants to do work like this, come showing you are capable of utilizing the knowledge about to be dispensed, and don't get angry if you don't show you are capable and no one helps. It is also for your benefit if you don't. Safety first!

Now back to the program in progress...

:)
 
  • Like
Reactions: Element Zero, abarson, Geri O and 2 others
#11
That's pretty high (assuming these are the driver transistors you are calling Q6&7)

There should be a few volts between emitters, but the top emitter should be slightly positive and the bottom emitter should be slightly negative with respect to ground.
 
  • Like
Reactions: DJ Bebop
#13
Those numbers look much better. The spread is the sum of the gate threshold voltages.

I think you have the currents reversed on the MOSFETs. If 3.7 means 3.7mV across the .33 ohm source resistors anyway.
 
  • Like
Reactions: DJ Bebop and Geri O
#16
15-20mV average on each bank. But I figured I would take care of the TP10 problem, before upping the bias.
I was afraid that bias could increase when replacing other components, and to be on the safe side keep it a little low. I've heard about fireworks when the bias gets too high.
 

Latest posts

Top Bottom
Back