PYE Black Box (PCL83) With Multiple Issues

[OldTechFan96]

Originally Posted by Gabe001

Try taking the treble control and c4 out of circuit and connect anode of V1a to grid of v1b directly.

I just jumped across the treble control and connected the anode to the grid and nothing happens.

Jumping across the control seems to kill the input signal.

[Gabe001]

If you replaced c4 with the correct value (worth re-checking, easy to make a mistake with the tiny print on ceramics) you may have to scope the entire path connecting v1a anode to v1b grid. Either you have a path to ground there or quite a bit of negative feedback is being introduced somehow. That's all I can think of. Quite a challenge you've got there

[OldTechFan96]

I've just done a few resistance checks and I don't think I'm losing signal to ground.

I had a good read of Valvepower's thread and It got me thinking about R30. The manual specifies that it is 4.75'' in length, but what I have identified as R30 must be half of what the service manual states.

When powered up, the voltage measured across R30 was 0V which doesn't seem right.

I lifted one side and measured its resistance with a Peak LCR meter. After zeroing the probes the measured resistance of R30 hovered between 0.01Ω and 0Ω.

I'm not too sure what to make of this. Is it an interesting line of enquiry or a red herring?

[PJL]

Red herring...

Lack of power could be one half of PP pair is not working so we need some AC measurements.

Be wary of the load from your scope probe as capacitance could kill your V1b grid measurement. It's also probably best to avoid grids as connecting the probe might cause instability. Are you giving us peak to peak as measured on the scope? Are you definitely measuring a signal from the source frequency at all times (i.e. not 50Hz or spurious oscillation)? I suggest add a picture of your scope image including the timebase setting.

Lets have AC volts on V1a anode, V1b anode, V1b cathode, V2a anode, V3a anode.

[OldTechFan96]

Last night I reinstalled the previously fitted audio output transformer because the rewound one did not change anything. After trying two original transformers and rewind, I'm sure that the problem is not transformer related.

AC Measurements for PJL

Scope: Hameg HM203-4
Audio Generator: Farnell LF1 Sine/Square Generator
Wattmeter: Marconi TF893A

The scope measurements were taken with the amp producing 1W into 1.5Ω with an audio signal of 2V P to P. The audio frequency was 1000Hz.

TB: 0.2ms/cm

Probe set to X10

V1A Anode: 2 3/4 div--> 50mV/cm--> 1.38V P to P.

V1B Anode: 2 div--> 50mV/cm--> 1V P to P.

V1B Cathode: 2 div--> 50mV/cm--> 1V P to P.

V2A Anode: 5 div--> 0.2V/cm--> 10V P to P.

V3A Anode: 6 div--> 0.2V/cm--> 12V P to P.

I hope this is useful!

[Valvepower]

Before I go any further and in pursuance of getting a basic input sensitivity and power output for the unit I have here I just measured input sensitivity and output power.

Firstly, as some background…, I’ve fitted a silicon bridge comprising 4x BY133. I have fitted a series resistor between the transformer secondary and bridge to reduce the elevated HT due to the use of a silicon bridge. This is currently 100Ω, however, I feel it should be slightly higher as personally the HT is still on the high side at 243V, but the 100Ω WW was all I had with a suitable power rating.

This unit has the standard wire ‘link’ in place of R30 – as I found it – and it matches the unit in question.

The PYE BB is connected to a 1.5Ω load, signal generator and oscilloscope and I’m running it at 1KHz with the all the controls at maximum and the output is just on the onset of clip (see attached scope trace) which corresponds to 3.1V and 6.4W.

I’m feeding in 1.3V (at 1KHz) into the (Co-axial socket from PU) input socket, and the voltage on the slider of R3 and junction of and R8 (100K) is 225mV - this what I’d call the basic input sensitivity of the amplifier circuit.

The input sensitivity seems on the low side compared to what I've seen elsewhere in this thread, however, to the best of my knowledge, the unit I have here is working OK, so the measurements I've made are 'as-is' and offered as such.

If it’s still of interest I can make the same measurements as in the previous post that's assuming the measurements of my unit are still sought? If so, I’ll endeavour do this sometime in the next few days, however, I’ve a few pressing 'personal' issues to attend to.

Terry

[PJL]

Look OK to me but how are you measuring watts? Certainly better than earlier measurements of V1A grid/V1b grid.

[OldTechFan96]

Watts are being measured with the wattmeter across the speaker terminals (speakers disconnected) and the meter set to the 10W range. The impedance is set to 1.5Ω.

When I measure the sensitivity of an amp I set the tone controls to the mid point, set the volume to max and then increase the level of a 1000Hz tone until the onset of clipping. Then I measure the AC volatge output the signal generator with a DMM. Will do this again to double check it.

Doesn't the low difference between what is measured on V1A Anode and V1B Anode suggest that there is a lack of signal or amplification between to two? Surely what's measured on V1B should be higher than V1A?

Thanks for the measurements Valvepower. Would you please measure the signal at V1A Anode and V1B Anode?

[OldTechFan96]

I just measured the resistance between V1A Anode and V1B Anode and my meter reads 157KΩ. The measurement does not change with the position of the volume and tone controls. Nor does it change with or without the valve removed!

[Valvepower]

Will do for V1 measurements tone controls at max. I’ll take them as per your output levels at 1KHz. I make 1W into 1.5Ω = 1.225V RMS...

As regards tone controls, normally if it were a Baxandall or passive circuit such used in the Mullard valve preamplifiers and suchlike I’d set mid position or adjust for best square wave at 1KHz, but with something like the BB I tend set to maximum to make basic measurements – it would be different if I were doing a frequency response though.

But saying that…, it was trying to get the best square wave at 1KHz by setting the tone controls that sent me down the PYE BB rabbit hole in the first place – as per my earlier thread and noting differences between the circuit of the unit I have and the one in the manual.

Terry.

[Valvepower]

Hello,

I’ve had a chance to get to bench and quickly measure the AC voltages using the Fluke 179 DVM.

1KHz and volume and tone controls at max

I’ve attached a [very] scribbled set of results – hopefully your able to read my scribble. I did two photos as I was getting shadows from the iPhone and lights.

Anyway, something to be getting on with…

Regards
Terry

[OldTechFan96]

Thanks again Valvepower your measurements have been helpful. It's good to know that your V1A Anode and V1B Anode measurements are similar.

I thought I'd see what output I'd get If I'd turned my signal generator all the way up. It's a bit distorted, but I do get 6.25W!

I think I've done all I can at this point so I'll try it back in the cabinet and see how it sounds.

[Valvepower]

Glad to read the measurements were of use – it makes it worthwhile.

I was looking at the overall gain from V1a grid to the speaker output and the values of the negative feedback components. Making oodles of assumptions both with the value of the open loop gain, positive feedback and the effect of the tone control EQ etc., the input sensitivity will be 92mV on V1a grid for a voltage of 1.22V on the speaker output. This is ideal so in practice it could be some millivolts out though.

I measured for 1.22V output 80.6mV on the junction of R3 slider and R8 (100K) which isn’t that far off the calculated 92mV.

looking at the anodes of V1a and V1b and assuming the treble control is set so the slider connected to V1b grid is on the track end which is connected to the anode of V1a (treble maximum) there will little difference between the AC voltage of two anodes V1a/b and also V1b cathode because V1b is a cathodyne/concertina phase splitter which has unity gain, in fact the level between the input and outputs of V1b could be very slightly less.

Terry.

[OldTechFan96]

I've not long finished getting the Black Box back together and I must say that it is sounding good and is more than loud enough with an Acos GP91-1SC.

I think I've spent a long time chasing a fault that did not exist.