Thanks for chiming in Perry.
R909 reads 2.345 Kohms and the resistance between the battery ground and non-bridging terminals is almost exactly the same.
I didn't have any source plugged in whilst making these measurements or the power supply measurements on the previous thread.
It was in fact your prompting about using DC coupling on that other thread which made me find this, I was looking for noise on the DC rails and realised they were different in magnitude with the DC coupling on.
Last edited:
Apologies Osvaldo, I was not very clear.
the original sets of measurements, the ones showing differing rail voltages, were referenced to the battery or primary ground.
The measurements I referenced in post #12 were taken using the speaker terminal grounds as oer your request in post #11.
Taken with my father's 'scope yesterday and referenced to battery ground. (but AC coupled, if that makes any difference)
When you are measuring anything on the secondary, it's best to ground the scope (or black meter probe) to the secondary ground. The only time that this can be a problem (rare) is if you have a shorted transformer (primary shorted to secondary). This is typically obvious because the secondary ground will be precisely the B+ voltage above ground.
For amps like this where the secondary is essentially isolated, the RCA shields of a standard signal source (like a head unit) will ground the secondary and bring it to 0v DC (unless the shield ground is damaged), assuming that the head unit is connected to the same supply as the amp.
The 2.7k resistor should be pulling the secondary to 0v on its own but sometimes there is a protection circuit or something similar that can pull it off it 0v. I don't see anything like that in this amp so you may have a leaking (electrically) insulator or some sort of contamination causing the offset.
For amps like this where the secondary is essentially isolated, the RCA shields of a standard signal source (like a head unit) will ground the secondary and bring it to 0v DC (unless the shield ground is damaged), assuming that the head unit is connected to the same supply as the amp.
The 2.7k resistor should be pulling the secondary to 0v on its own but sometimes there is a protection circuit or something similar that can pull it off it 0v. I don't see anything like that in this amp so you may have a leaking (electrically) insulator or some sort of contamination causing the offset.
Recheck this measurement DC coupled, and with the ground crocodile to the audio ground.
It is strange to me that having equal voltages, be there different amplitude in the rectangular wave. Something smells me not good.
It is strange to me that having equal voltages, be there different amplitude in the rectangular wave. Something smells me not good.
I believe that the waveforms appear offset only because the entire secondary is offset. Grounding the scope to the secondary ground (you can use the non-bridging speaker terminals as ground) should give symmetrical waveforms.
I agree. In my opinion the amp is OK and the guy is making any mistake taking measurements. All we did it once in our lives, no?
Thanks Perry, that makes sense.
The board itself looks fine, do you think it could be the capcitor that seems warmer than the other?
It heats up even at idle and with no signal in.
It's C909.
Try measuring with the oscillo across C912 and R909. Perhaps the isolating inside the traffo is degraded and is passing some primary leak to the secondary, trough the cap. This cap must be at room temperature. There is no AC current flowing in it of such importance to became it hotter than ambient.
I measure 4.8 volts across C912 and R909, I struggled to get a measurement with the 'scope, I think my probe ground is too coarse to grip the tiny resistor properly.
The RCA shields are also at the same potential with reference to the battery negative.
The RCA shields are also at the same potential with reference to the battery negative.
Last edited:
If you're referring to C912, clip a jumper between the primary ground and either of the non-bridging speaker terminals. The voltage across that cap will drop to 0v. Does it still heat up?
The cap that heats up is the main rail cap, c909.
It’s gets hot whereas it’s counterpart does not.
C912 doesn’t seem to get hot.
It’s gets hot whereas it’s counterpart does not.
C912 doesn’t seem to get hot.
With your scope ground connected to one of the non-bridging speaker terminals and the vertical amp on the scope set to 10v/div. Do you see a perfectly clean, straignt line on the negative terminal of C909 and the positive terminal of C908?
I have clean, straight +- 32 v at those points, albeit not under any load.
Having taken the board up from the heat sink to take the measurements I think I might have found a dry joint at the centre leg of D903. I'm going to try and re-solder it to is if it just that.
Thanks for your help Perry.
Edit: Re-soldering the diode leg did not fix the issue.
Having taken the board up from the heat sink to take the measurements I think I might have found a dry joint at the centre leg of D903. I'm going to try and re-solder it to is if it just that.
Thanks for your help Perry.
Edit: Re-soldering the diode leg did not fix the issue.
Last edited:
If the capacitor is heating up with no excessive ripple (clean DC) and there is no excessive DC, I'd suspect that the cap is leaking (electrically).
I see Perry, thank you.
They are an odd shape ( 30 D x 25 H ) 3900uF 40v.
I assume it would be ok to replace them with slightly higher value 50v caps?
They are an odd shape ( 30 D x 25 H ) 3900uF 40v.
I assume it would be ok to replace them with slightly higher value 50v caps?
Mouser appear to have some but of a 85c persuasion. It seems the choice at 4700uf is a bit better. I shall order and fit them.