AB165 Bassman B+ voltage issues

[old tube]

Hi recently got an early 1968 Bassman as a fixer upper, though it does work. Circuit has been AB165 but it's had the usual mods to NFB and has proper bias adjustment.

I'm in the UK. The amp has originally been a USA non-export model and the transformer is not original ... but I don't know if it was changed out because it died or just to get running at 240V without a second transformer.

There is an issue with the transformer as it has no alternate mains selection and it's probably a 220/230V model. I have a wall voltage here of 245V, the secondary is putting out 350V to the rectifier diodes. So I'd expect around 500V (too high, I know) coming off the diodes. But here's the thing, the voltage after the diodes is 554V and at the first smoothing caps is 277V per side suggesting that there isn't a ground reference issue with the caps. The extra voltage is seen at all the B+ nodes.

I'm guessing therefore that the issue is with the diodes, passing AC? However, the amp doesn't hum! I've been advised that if I attached a high voltage cap (1000V) to the diode out, having disconnected the wire to the smoothing caps, that if the diodes are leaking AC I'll see it across the capacitor. So assuming that is the case, do I need to replace all the diodes? or is there a way of finding the culprit (s)

Other perhaps pertinent information is that the Transformer vibrates badly when it is on, switching to standby it stops and is quiet. It doesn't get overly hot even after 30 minutes or so. The amp works though it doesn't sound as it should; volume is there but pretty thin tone and quite a high level of hissing /electrical noise. It doesn't hum noticeably however, which I would have expected if there was AC passing the diodes.

If anyone has any other ideas please don't hold back.

I've got a new, proper spec transformer for it, but want to sort this issue before I install it.

[Stevem]

Are you sure your meter is reading right, as 350 volts will leave you with 490 volts after the diodes at the first filter node?
What plate B+ voltage do you read at the power supply side of the two 100k resistors for V1?

[pdf64]

the voltage after the diodes is 554V and at the first smoothing caps is 277V per side

http://schems.com/schematicheaven.net/m ... _schem.pdf
The 2 reservoir caps are in series, so the full VB+ won't appear across each individually.
Are you mixing up readings in standby and operate modes?
Best to stick to 'operate' mode, no signal, aka quiescent or static.
What's the voltage across the heaters, eg 6.3Vac?

[old tube]

Thanks for your replies guys.

Ok, so I've put a new battery in my meter and re-taken voltages with standby off as suggested (I had measured with standby on before)

The first thing I noticed is that if I measure the AC from the PT at the start of each diode string I get 276V at the lead for the first string and 370V on the other! I'm sure both were the same the other day when I measured them. I've checked them a few times turning off and on between and it's consistent.

The heaters are spot on at 3.15 each side, suggesting that the PT is a 240V.

The DC on the B+ chain measured at the board in each case is;

500V, 490V, 461V, 357V

Is the different voltages coming from each HT output wire of the PT the clue here?

[David Root]

In play mode they should be the same within a few volts. On standby they can do what you are observing. Are you sure you measured in play mode? The DC chain voltages seem right given your wall mains voltage.

[Phil_S]

Ok, so I've put a new battery in my meter and re-taken voltages with standby off as suggested (I had measured with standby on before)

On terminology, in the interest of clarity, I suggest you indicate the switch is closed (operating or on postion) or open (off position.) I find open and closed to be clear. I find off standby or on standby impossible to understand. They seem to be used interchangeably.

The first thing I noticed is that if I measure the AC from the PT at the start of each diode string I get 276V at the lead for the first string and 370V on the other! I'm sure both were the same the other day when I measured them. I've checked them a few times turning off and on between and it's consistent.

This seems all wrong. Either there is something wrong with the PT, the circuit, the meter, or the way you are doing it. Please describe exactly what you are doing and take the measurement with clips, not hand held probes. Clips will stabilize the reading. Hand held readings can sometimes be iffy due to poor or intermittent contact. Clip the back to ground and the red to where the PT wire meets the first diode, once on each side. If you get the same with clips, then we need to look for something else. Please excuse me if you are already doing this.

If you really have 276-370, I'd lift the PT leads from the diodes and see what each leg reads without any load. If it remains the same, you know it is the PT. At that point, I'd get another PT.

The heaters are spot on at 3.15 each side, suggesting that the PT is a 240V.

Measure heaters by clipping one lead to one filament wire and the other lead to the other filament wire. Your meter should indicate 6.3VAC. Don't measure each lead to ground. It sounds like you are doing that.

The DC on the B+ chain measured at the board in each case is;
500V, 490V, 461V, 357V Is the different voltages coming from each HT output wire of the PT the clue here?

If you are really seeing 500VDC at the first B+ note (center tap of output transformer) that suggests input to the diodes is 714VAC. This is not consistent with your report of 276 and 370, which is 646V. 714VAC is reasonably consistent with your first report of 350-350. Backsolving for input voltage says its 357-357.

The B+ ladder is linear. Once voltage is set at the first node, the rest follow based on the circuit.

[pdf64]

I measure the AC from the PT at the start of each diode string I get 276V at the lead for the first string and 370V on the other

Maybe one of the rectifier diodes is open circuit?

Is there a bias tap?

If so, maybe it's been mixed up with the CT?

[old tube]

Many thanks for all the advice, much appreciated.

So I've re-measured using croc clips and the standby switch closed (operating) and allowing a few seconds for readings to stabilise.

Across both HT wires before diodes 670V
HT supply wire 1 388V
HT supply wire 2 282V

Bias tap 53V

Heaters 6.3V

B+ after diodes 508V

Assuming next move would be to disconnect the HT supply wires from the diode strings and measure the PT direct?

[Phil_S]

Assuming next move would be to disconnect the HT supply wires from the diode strings and measure the PT direct?

That's what I'd do. If it shows reasonably symmetrical voltage, I'd suspect the diodes. if your meter has a diode checking function, you might do that before you disconnect the PT. If I found one bad one, I'd replace all just for good measure. Diodes are cheap. You can put in a pair of 1N4007 in place of the triple strings (though you loose the minor voltage drop) and it will work just fine.

It still doesn't make sense that you see 508VDC. Divide that by .707 = ~720VAC for the input. I still think there is something wrong about the voltage measure on the leg that shows you 282 or 278.

I wonder, just for grins, can you get a VDC reading between each of the diode junctions -- that's on either side of the middle diode to ground.

[Stevem]

Besides the NF mod these Amps have a bad sounding PI section in and of itself so that would be the next area to attack after you resolve this possible B+ issue!

[old tube]

Ok so taken the PT HT leads from the diode string and measured direct to the HT section centre tap.

Lead 1 390V
Lead 2 280V

Consistent at least! Looks like it is the HT section of the PT at fault?

The PT does vibrate horribly when the amp is fully on, tbh I'd expect smoke at any time! I've got a proper spec replacement for it and will be installing it later this week.

Sorry Phil_S, I forgot to check the diode string as you suggested before I disconnected it all. Guess I'll find out if there is an issue there as well when I install the new transformer.

[old tube]

Besides the NF mod these Amps have a bad sounding PI section in and of itself so that would be the next area to attack after you resolve this possible B+ issue!

Someone has changed some of the PI values already, but to be honest the amp just sounds nasty, most likely through the B+ problem. In this example, the 22k resistor has been changed to 10k, the coupling cap into the PI reduced to 0.01uF (!).

There are a few resistors around the circuit of the wrong decimal value .. someone couldn't read 4 line resistor codes! They also didn't know the difference between log and linear pots as the normal channel volume is linear, they would have been as well installing a switch for full on or off!

The NFB is currently 4.7k coming from the OT output with a 470k ground reference! The cap original to the circuit has been removed. I intend to use 820/100 ohm pair for the NFB.

Output coupling caps are 0.1uF which I plan to replace with 0.05uF in the first instance.

Any recommended changes to the PI circuit much appreciated. There is a lot on-line and of course everyone says their mods are best!

[pdf64]

Ok so taken the PT HT leads from the diode string and measured direct to the HT section centre tap.

Lead 1 390V
Lead 2 280V

So is there a bias tap on the HT winding, as well as the centre tap?
Might they've been connected the wrong way around?

[Phil_S]

Interesting thought about a bias feed. How many wires and what color are they? A picture maybe? Otherwise, I'd tend to agree that PT is ready to retire.

[old tube]

Ok so taken the PT HT leads from the diode string and measured direct to the HT section centre tap.

Lead 1 390V
Lead 2 280V

So is there a bias tap on the HT winding, as well as the centre tap?
Might they've been connected the wrong way around?

The bias voltage is reading 53V AC before the 470 ohm resistor and diode. which seems right to me. Would it not be a lot higher if the wires were reversed? as the Bias tap would be mid point of the secondary? Is it safe to reverse them to "find out"?

It's a no-name transformer and has minimal wires in and out. On the output side, there are two red wires for the HT, a blue wire comes out right beside them and is currently to ground. there is a yellow wire further to the right which goes to the bias feed and then further again right, two green wires going to the heaters, no centre tap for the heaters.

Hmmm ... just done some googling and the classic colour code for HT centre tap is apparently red/yellow. Should I just reverse the yellow and blue leads?