Another KCS47 Restore...No High Voltage/B+ Down 25V

[Chris K]

Starting fresh with an unrestored chassis. I figured I'd start in the power section like everyone else does but you need to make some rigging to be able to test and adjust it.

I took a tip from Kevin and built this rig for the 5" electrostatic test CRT.



I do not have the speaker out of the cabinet and therefore, I needed to make a jumper. The 375V rail goes into the sound section and anywhere else the 375V needs to go. The jumper is on the speaker so without it, I jumped it on the speaker connector.




Where the DMM is connected reads 350V and not the 375V in the schematic. The B+ on pin 8 of the 5U4 reads around 25V lower than the approximately 387V in the service data. I did some recapping in the power section and all three electrolytic caps have been replaced with adapt-a-cap substitutes. I covered them with some cardboard tubes I had laying around but they are grounded to the chassis.

I rigged a HV cable from the metal cone energizing socket on the HV cage. I replaced the 4 film caps in this section. The flyback is a Merit direct replacement for the original one that had an open section.




I've switched out the HV rectifier, horizontal output and damper tubes numerous times with known good tubes. I get about 3KV on the anode output when I test it at the socket.

My guess is the low HV is a function of the low B+ but I'm not sure of that. I've tried several 5U4s. If this is a problem with a resistor in the voltage divider, why would the voltage be low on pin 8 of the 5U4?

Anyway, below is the section giving me difficulty. I've made no other changes in the chassis other than the capacitor replacements I listed above.

[bandersen]

I would not consider your B+ low. Figure those stated voltages are +/- 20% or more. Plus you haven't recapped the whole set yet. Some circuits could be drawing more power than typical.
Measure the voltages on the 6BG6 and check the grid drive waveform. Likely your hor. freq. is off.
I wouldn't expect you to get full HV until the whole hor. oscillator and output circuit is gone over.

[Chris K]

Okay thank you I’ll do that and report back

[Chris K]

The horizontal frequency after a slight adjustment looks right but the waveform is noisy.



Pin 8 is 293V DC Should be +329V DC
Pin 5 is -4.8V DC Should be -33V DC
Pin 3 is +11V DC Should be =7.2V DC

HV is around 3.8KV. The waveform grid voltage is way off

[Kevin Kuehn]

If you haven't replaced the leaky paper caps in the horizontal circuit, you need to. Same with the vertical. 3.8 kv means it's basically trying to work. But if there's leakage it's messing up the bias on the 6BG6. With so little grid bias the 6BG6 may be pulling a hefty amount of plate current. 110ma according to your voltage reading on pin 3.

[Chris K]

Ok I’ll finish replacing the caps and check the resistors in both deflection circuits completely and see what that does.

[Chris K]

I think what I’ll do Kevin is post a list of the caps I’ve replaced and a bigger section of the schematic. I’d like to know for education purposes what’s causing this specifically and how I can diagnose issues like this in the future instead of doing a bulk recap in the power and deflection sections. I’d like to be able to say oh that’s what caused it and not well I changed everything and that took care of it.

[bandersen]

I get what you are trying to do, but it's been my experience it can get frustrating.

All of the paper caps are leaky.
Many of the electrolytics will be excessively leaky or open.
Many of the resistors will be high.

It's not like troubleshooting decades ago where most of the parts are fine.

[Chris K]

Point taken. Thanks!

[Kevin Kuehn]

Originally Posted by Chris K

I think what I’ll do Kevin is post a list of the caps I’ve replaced and a bigger section of the schematic. I’d like to know for education purposes what’s causing this specifically and how I can diagnose issues like this in the future instead of doing a bulk recap in the power and deflection sections. I’d like to be able to say oh that’s what caused it and not well I changed everything and that took care of it.

You can do that but you'll be spending a lot more time getting things corrected. Right now you know the grid of the horizontal output tube is under biased, and it's pulling more plate current than intended. Besides paper caps it's possible the 560pf mica grid coupling cap is leaking positive voltage from the horz oscillator onto the horz output grid. You could try pulling the 6BG6 and see there's any residual voltage remaining on pin 5 of it's socket. Speculating we can assume the higher than normal plate current the 6BG6 is drawing could be pulling a B+ down few volts. That happens because these old sets don't have much if any voltage regulation. And the horz out is probably the biggest consumption of power.

[Chris K]

Originally Posted by Kevin Kuehn

You can do that but you'll be spending a lot more time getting things corrected. Right now you know the grid of the horizontal output tube is under biased, and it's pulling more plate current than intended. Besides paper caps it's possible the 560pf mica grid coupling cap is leaking positive voltage from the horz oscillator onto the horz output grid. You could try pulling the 6BG6 and see there's any residual voltage remaining on pin 5 of it's socket. Speculating we can assume the higher than normal plate current the 6BG6 is drawing could be pulling a B+ down few volts. That happens because these old sets don't have much if any voltage regulation. And the horz out is probably the biggest consumption of power.

Thanks as always Kevin. Do you mind expanding on this a bit for me? I thought bias voltage just turned a tube on and off. There's a ramped relationship between bias on the grid and the plate output? Is a biasing voltage always a negative voltage? Since -4.8V is a higher voltage than -33 volts DC does that account for the higher current on the plate? Confused as usual by the details. Can you explain to me how a 6BG6 works other than the filament voltage?

[Kevin Kuehn]

Originally Posted by Chris K

Thanks as always Kevin. Do you mind expanding on this a bit for me? I thought bias voltage just turned a tube on and off. There's a ramped relationship between bias on the grid and the plate output? Is a biasing voltage always a negative voltage? Since -4.8V is a higher voltage than -33 volts DC does that account for the higher current on the plate? Confused as usual by the details. Can you explain to me how a 6BG6 works other than the filament voltage?

You've pretty much got it. The negative bias on the grid(relative to cathode)sets the idle current, or operational point. So in some cases it can be turning on and off, like a switch, and other times it's partially on in neutral, so to speak. As far as I know normal receiving tubes use negative bias on the grid. If that's not clear I'll attempt to elaborate.

[Kevin Kuehn]

If you look in a technical text they use the term Quiescent Current. Different classes of amplifiers idle at different currents. Class A for example is turned on fairly hard at idle, vs the horizontal output is biased closer to cutoff, and acts more so like a switch.

[Electronic M]

Tubes don't just amplify AC and aren't just on and off. The plate current, with no signal, in anything with a control grid (other than a thyratron) is proportional to the DC grid bias. The exact bias voltage to achieve these following these 3 states varies with the tube type number and wear, but here goes. A tube with strong negative grid voltage (for instance -50V) is in cutoff and no plate current flows, a tube with almost no negative (or even positive grid voltage is in Saturation (maximum plate current it can physically pass... sometimes in excess of ratings is flowing.. often hundreds of mA), and if a tube is biased somewhere between the two it's probably in the linear region where a fraction of a volt grid change results in a mA change in plate current.... often the linear region is somewhat non-linear as it approaches the other regions.
In a single ended class A audio amp DC bias is critical to getting maximum volume...You want bias in the exact center of the linear region or the grid swing will clip either the tops or bottoms off the waveform before maximum achievable amplitude is reached.... Also audio output transformer DC saturation at high current can force reduction in plate current bias because the transformer hits saturation before the tube. In a push pull audio amp you want to avoid class A operation as you waste power and don't get maximum amplitude. In push pull the DC transformer saturation from each tube cancels the other and each tube can handle half the signal, but can be in cutoff for the other half without issue... However a little bit of zero signal plate current is necessary to avoid clipping from a messy handoff between tubes at the 0 crossing of the signal....If that distortion doesn't matter Class B can be used.

Horizontal sweep systems use class C where the load has resonance in the desired frequency range (like a bell) and the tube only conducts for a small portion of the AC signal sufficient to ring bell, keep it ringing and at the right ring strength.... Horizontal output systems Instead of using straight forward DC bias circuits that resistively divide the DC supply get their DC bias power from the signal. They use the grid leak detector principle where grid-cathode rectification of the AC signal creates a DC bias that regulates tube current.

Triodes, terrodes and penthouse act the same as signal amplifiers with regard to DC and AC control grid behavior in single input amplifying or switching applications...The only thing the extra grids do is change the slope of the linear region relationship between grid voltage and plate current...They increase gain which is useful enough to merit the additional cost and complexity. With extra grids you can achieve 2 input amplification or switching which can be seen in color TV color demodulation and some gated sync separator/AGC circuits respectively... Superheterodyne converter and or mixer stages also use this... You shouldn't see this in a horizontal output circuit unless it's the infamous single tube horizontal Muntz circuit in that one model they made...

[Chris K]

Originally Posted by Kevin Kuehn

You can do that but you'll be spending a lot more time getting things corrected. Right now you know the grid of the horizontal output tube is under biased, and it's pulling more plate current than intended. Besides paper caps it's possible the 560pf mica grid coupling cap is leaking positive voltage from the horz oscillator onto the horz output grid. You could try pulling the 6BG6 and see there's any residual voltage remaining on pin 5 of it's socket. Speculating we can assume the higher than normal plate current the 6BG6 is drawing could be pulling a B+ down few volts. That happens because these old sets don't have much if any voltage regulation. And the horz out is probably the biggest consumption of power.

Pulled the 6BG6 and there's no voltage on the grid (5) and the plate (8) is 410V DC