Meck XB-702 restoration

[Kevin Kuehn]

Originally Posted by bandersen

Interesting results with a 1M cathode resistor. The gain does drop, but the grid is far less sensitive to external noise. However, there is still a buzz present. Grounding the grid does nothing to reduce it. Grounding the cathode reduces it by about 50%.

I'll continue to experiment.

1meg is way to much for a cathode bias resistor, as there's only 390k in the plate. If you look in the back of an RCA tube manual they'll give recommended values for resistance coupled amplifies, and I'm pretty sure you'll find the 6AT6 included in one of the charts. You may also reduce the value of the plate resistance. The idea being to reduces both the impedance and gain of the stage. Just a guess would be about 5K in the cathode if you leave the plate resistor alone.

[bandersen]

Oh, right of course. Jeez I misread your earlier post where you mentioned adding an unbypassed cathode resistor and reducing the grid to 1Meg. I'm surprised it works at all!

[Josef]

As a foreign guy I cannot say a lot to technical details of this set.
Anyway: your work on restuffing the caps is more like I would do on prewar or ultra rare sets

[bandersen]

I did find the RC coupled amp section in the RCA tube manual and have been experimenting. I currently have an 8.2K cathode resistor and 1M grid resistor. The hum/noise has definitely been reduced with no noticeable drop in gain.

Will try smaller values of cathode resistor and grid resistor.

Also experimented with a cathode bypass cap. Adding one to the circuit seems to increase the noise.

Tried shorting out the 0.002 input cap too. Not much difference either way.

[Kevin Kuehn]

I think you'll find increasing the cathode resister will reduce gain even more, but there is a limit unless you reduce the plate resistor to keep the overall impedance. With the plate and cathode resistor equal you basically have equal and out of phase signals from the plate and cathode respectively(as in a phase splitter), but with about zero gain. I've always disliked grid leak bias on the first audio in a radio for this same reason. Besides being buzzy they distort easily.

[bandersen]

I think I found a design the works well. I increased the coupling cap from .002 to .01, decrease the grid resistor to 470K and added a 4.7K plus 2.2uF to the cathode.

The only noise left is the occasional buzz when there is harsh contrast on the screen. The slight reduction in gain is a good tradeoff I think

[bandersen]

Originally Posted by Josef

As a foreign guy I cannot say a lot to technical details of this set.
Anyway: your work on restuffing the caps is more like I would do on prewar or ultra rare sets

Thanks. I don't know if this model qualifies as ultra-rare but it's certainly uncommon. Especially with so many original components.

[Kevin Kuehn]

Have to wonder what they were thinking using grid leak bias in this situation without the diodes. I suppose trying to save one resistor and maybe a cap.

[bandersen]

That's all I can figure. I'm also thinking that might be the reason the speaker was moved to the opposite site. An attempt to reduce noise or feedback.

The set seems happy with the changes I made

[bandersen]

The only remaining issue is some twerking. I suppose some poking around with a scope is the best approach. I'll start with the deflection waveforms.

[AlanInSitges]

I've been watching this for almost 20 minutes and haven't seen any twerking at all. Good job!

[Electronic M]

Originally Posted by AlanInSitges

I've been watching this for almost 20 minutes and haven't seen any twerking at all. Good job!

[init4fun]

Originally Posted by AlanInSitges

I've been watching this for almost 20 minutes and haven't seen any twerking at all. Good job!

Ya didn't miss much, twerking was so 2000's .....

[old_coot88]

Taint funny at all.

[bandersen]

You guys are too much. Here's a short clip showing the "twerk" in action: https://youtu.be/SfvSbQZUtVM