Adjusting an old bias pot did some damage and I am unsure how to proceed

The bulb was glowing quite well with variac at 60V so that was when I took the measurments. Probably not necessary to list them all but I'll do so anyway. This is how they are arranged looking at the component side of the board.

R316 1.1mv R317 0.0mv R418 0.0mv R419 2.1mv

R317 492mv R316 261mv R416 0.0mv R417 0.0mv

R317 is associated with Q307 and R316 with Q308. I assume it is those two Q's which are goners.

EDIT: FWIW, I just ordered some metal film 1/2W 820 ohm resistors fo replace those two cracked ones near the bridge rectifier.
 
Member
Joined 2021
Paid Member
The bulb was glowing quite well with variac at 60V so that was when I took the measurments. Probably not necessary to list them all but I'll do so anyway. This is how they are arranged looking at the component side of the board.

R316 1.1mv R317 0.0mv R418 0.0mv R419 2.1mv

R317 492mv R316 261mv R416 0.0mv R417 0.0mv

R317 is associated with Q307 and R316 with Q308. I assume it is those two Q's which are goners.

EDIT: FWIW, I just ordered some metal film 1/2W 820 ohm resistors fo replace those two cracked ones near the bridge rectifier.
I would use a bulb minimum 100w,that way you should be able to work on it without blowing the fuses, and do one channel at a time
 
So back to where we were. If the bias adjusts correctly for both channels (remember final check is on full mains) then it should all be working. Make sure the offset is still zero before connecting speakers.

Yes, it would appear so - back to where it was.

This morning I left the amp powered up for an hour or so. It was plugged directly into wall power, no variac and no DBT.

The bias current I was able to measure was about the same for both channels. The lowest voltage achievable was about 38mv and could be adjusted but the highest voltage achievable was about 48mv.

I had hoped that bias current would begin much lower than 38mv and adjust higher than 48mv.

With bias pots at minimum, DC offset at speaker terminals is 37mv on the right channel an 3mv on the right.
 
Administrator
Joined 2007
Paid Member
The bias current I was able to measure was about the same for both channels. The lowest voltage achievable was about 38mv and could be adjusted but the highest voltage achievable was about 48mv.

I had hoped that bias current would begin much lower than 38mv and adjust higher than 48mv.

You can tweak that resistor a bit more. Look at the image in post #57. You need to go a little higher still to say 2k7.

As an alternative you can also add a resistor in parallel with R309 which is the 5k6 feeding the top of the preset which will lower the overall value. A 22k would be a good starting point.

Only do one or the other, not both. Unfortunately it is trial and error to get the bias in range.
 
You can tweak that resistor a bit more. Look at the image in post #57. You need to go a little higher still to say 2k7.

As an alternative you can also add a resistor in parallel with R309 which is the 5k6 feeding the top of the preset which will lower the overall value. A 22k would be a good starting point.

Only do one or the other, not both. Unfortunately it is trial and error to get the bias in range.

I opted for adding a 22K resistor in parallel with R309. After the amp was idling for over 1/2 hour, bias was 37mv with the pot at minimum. The other channel has not had a 22K resistor added and bias on that channel was exactly the same, 37mv.

The additional resistor made no difference.

After removing the 33k ohm paralleld resistor, I can try upping that other resistor to a 2.7k to see what happens. The reason I chose to install a resistor in parallel was because I was a bit aprehensive swapping resistors due solder pad fragility. It does not take much to lift them.

But as mentioned, I will take that step but not til tomorrow. It's late now where I am.

May not be much help, but I have had this loads of times with the Nads, and it's almost always been a bit of broken print, and thorough continuity checks normaly found the issue

This might be something to consider. The traces, solder pads especially, do not tollerate much reworking.
 
Administrator
Joined 2007
Paid Member
I opted for adding a 22K resistor in parallel with R309. After the amp was idling for over 1/2 hour, bias was 37mv with the pot at minimum. The other channel has not had a 22K resistor added and bias on that channel was exactly the same, 37mv.

The additional resistor made no difference.

The theory is sound. Go lower in value with the parallel resistor such as 18k, 15k and 10k.
 
  • Like
Reactions: 1 user
Member
Joined 2021
Paid Member
I opted for adding a 22K resistor in parallel with R309. After the amp was idling for over 1/2 hour, bias was 37mv with the pot at minimum. The other channel has not had a 22K resistor added and bias on that channel was exactly the same, 37mv.

The additional resistor made no difference.

After removing the 33k ohm paralleld resistor, I can try upping that other resistor to a 2.7k to see what happens. The reason I chose to install a resistor in parallel was because I was a bit aprehensive swapping resistors due solder pad fragility. It does not take much to lift them.

But as mentioned, I will take that step but not til tomorrow. It's late now where I am.



This might be something to consider. The traces, solder pads especially, do not tollerate much reworking.
Seriously check because if you have done as much work as I think you have, that board will be complaining alot by now, and the cracks won't be obvious
 
  • Like
Reactions: 1 user
Another development arose. DC voltage has disappeared.

There is no DC voltage at the large storage caps, the pilot (power on) lamp does not light but both fuses are intact. This was without DBT and variac at full power. The variac's fuse is good and wall power is good.

This occurred after I unsoldered that 22K ohm parallel resistor at R309 but before replacing it with the 18K ohm I was intending to.

Shortly after I acquired this amp several years ago, I discovered there were faulty output transistors which I replaced. I also learned that the bridge rectifier in these models had a history of failure. At that time, I was told how to assess if the rectifer I had was good or not and it seemed to be OK.

The rectifier in question is identified as 5B2. I was concerned enought that I removed it from the board just now to check it once more. DMM in diode mode resulted in good readings so the rectifier looks fine.

EDIT: I have yet to search for faulty connections. But would not a total loss of power point to some component failure?
 
If you pull the mains plug out and measure between the two pins on the plug (live and neutral) on ohms on your meter it should read quite low, maybe not much more than 10 ohms. Remember to turn the amp mains switch on to check this to give continuity.

In post #66 was the suggestion to measure ohms across the blades of the AC power plug. With power switch on. I just repeated that proceedure and my meter displaed OL whereas before, it was 5.4 ohms.

There is continuity between the plug blades and the ends of the power cord wires inside the amplifier. I thought there was a remote possibility of a break in the power cord connection but continutiry would confirm the cord is good.

I measured AC voltage where the AC cord terminates at the accessory AC plugs on the amp's rear panel. My meter displayed 142 volts AC which seems high.

Could that have damaged the transformer?
 
Administrator
Joined 2007
Paid Member
Its unlikely you have damaged the transformer, it would have to get seriously hot to fail.

Always refer to the circuit and trace things through :) The accessory sockets are 'before' the mains switch so the switch is the next likely suspect. With the amp unplugged try and measure continuity directly across the primary. You should have continuity between Brown and Green and between Orange and White.

At this point the latest problem sounds more like the switch being open.

The AC voltage sounds very high if you have nominally 120 volt mains but a low battery in a meter can cause that effect.