Dumble Overdrive Special #124 build video series

[didit]

NO it's a low pass filter.
We are trying to give user experiences to Pompeii. He's got the balls to show every move he makes on YouTube, which I think is a modern kind of heroism. So let's not pollute this topic with misunderstandings.

I am not an electronics engineer. I just learned this stuff off the interwebz. I am not ashamed to say that, schools are dead, I learned to repair my toaster, pave my patio, and a lot more too. So I'm also begging to be corrected.
[...]
what's a snufa?

Hi-pass shunted to ground cuts — ie a true low-pass effectively. Now corrected more than once. Exactly.

Best .Ian

[erwin_ve]

Screenshot from 2018-03-14 23-33-10.png

Found it: I think it's a cap and then a resistor to ground, hence a high pass filter:
Screenshot from 2018-03-14 23-35-49.png
Screenshot from 2018-03-14 23-39-09.png

What am i missing?

NO it's a low pass filter.
We are trying to give user experiences to Pompeii. He's got the balls to show every move he makes on YouTube, which I think is a modern kind of heroism. So let's not pollute this topic with misunderstandings.

I am not an electronics engineer. I just learned this stuff off the interwebz. I am not ashamed to say that, schools are dead, I learned to repair my toaster, pave my patio, and a lot more too. So I'm also begging to be corrected.

If the signal goes through a capacitor and then through a resistor, it will create a high pass filter. It will cut out part of the signal on the bass side of the spectrum. Check the wiki link for the formula to calculate which frequencies will be affected. Instead of messing with that formula you can just plug your values in this calculator:
http://www.trance-cat.com/electrical-ci ... ulator.php

The graphs accompanying it shows that all with a 250pF and 1k5 it cuts anything below 424Hz. With 270pF, anything below 392Hz.
If you know your guitar low E string is 82Hz, it's clear we are affecting the guitar signal. As this part of the circuit is the distortion, this matters even more to the 'sound' of the distortion.
Putting a 500pF in there will make a substantial difference. Then we are doubling the value... But really saying 20pF more or less will make a difference... in distortion circuit maybe yes, but as I said, you need to have high precision capacitors and be able to measure them accurately. It's hard measuring, I think Phil wondered about this in one of his videos, because the capacitance of the leads need to be taken into account to measure values that low really accurately.
This is also a non-inverting feedback loop... it feeds signal back through the cathode. But I might be wrong...

The snubber cap is not in the signal path! So there's your wrong assumption. If you don't' believe me I would suggest you draw a schematic of a tone circuit as seen on guitar...
Also it's not a feedback loop. The highs are shunted to ground through the cathode bypass cap. In case of not having a cathode bypass cap and only having a cathode resistor I would agree with you.

[jam-mill]

[...]

I appreciate everyone's help here. I also was very happy that Charlie Wilson made it to the last livestream and was a great contributor to the discussion. I would be happy if any of you other geniuses would come along and forward the discussion and catch my SNAFU's

Phil, don't mean to hijack the thread, discussion is still on topic of the relevance of that cap beyond mere hearsay. I do think your a hero for what you are doing with your build, because despite a lot of mp3 demoes, there is in my perhaps too critical mind, too much doubt about the documentation on the Dumbles to start a build myself. I think anybody on this forum would be dying to see anybody just do an iPhone video display of the amp they've built. Inside and out. Run over the problems they've encountered and how they solved it. The biggest plug for this forum is still Fuchs mentioning it in his interview... that's the only reason I came here. That made me believe there is info here, but it really seems to be scattered.

what's a snufa?

SNAFU - Situation Normal, All Fouled Up

[pompeiisneaks]

Screenshot from 2018-03-14 23-33-10.png

Found it: I think it's a cap and then a resistor to ground, hence a high pass filter:
Screenshot from 2018-03-14 23-35-49.png
Screenshot from 2018-03-14 23-39-09.png

What am i missing?

NO it's a low pass filter.
We are trying to give user experiences to Pompeii. He's got the balls to show every move he makes on YouTube, which I think is a modern kind of heroism. So let's not pollute this topic with misunderstandings.

I am not an electronics engineer. I just learned this stuff off the interwebz. I am not ashamed to say that, schools are dead, I learned to repair my toaster, pave my patio, and a lot more too. So I'm also begging to be corrected.

If the signal goes through a capacitor and then through a resistor, it will create a high pass filter. It will cut out part of the signal on the bass side of the spectrum. Check the wiki link for the formula to calculate which frequencies will be affected. Instead of messing with that formula you can just plug your values in this calculator:
http://www.trance-cat.com/electrical-ci ... ulator.php

The graphs accompanying it shows that all with a 250pF and 1k5 it cuts anything below 424Hz. With 270pF, anything below 392Hz.
If you know your guitar low E string is 82Hz, it's clear we are affecting the guitar signal. As this part of the circuit is the distortion, this matters even more to the 'sound' of the distortion.
Putting a 500pF in there will make a substantial difference. Then we are doubling the value... But really saying 20pF more or less will make a difference... in distortion circuit maybe yes, but as I said, you need to have high precision capacitors and be able to measure them accurately. It's hard measuring, I think Phil wondered about this in one of his videos, because the capacitance of the leads need to be taken into account to measure values that low really accurately.

This is also a non-inverting feedback loop... it feeds signal back through the cathode. But I might be wrong...

As it appears to me, it's a negative feedback loop it sends signal directly from the anode (inverted from input) to the cathode (follows input) or vice-versa for a specific frequency due to that capacitor. anode and cathode are inverted and therefore would be negative feedback.

I appreciate everyone's help here. I also was very happy that Charlie Wilson made it to the last livestream and was a great contributor to the discussion. I would be happy if any of you other geniuses would come along and forward the discussion and catch my SNAFU's

Phil, don't mean to hijack the thread, discussion is still on topic of the relevance of that cap beyond mere hearsay. I do think your a hero for what you are doing with your build, because despite a lot of mp3 demoes, there is in my perhaps too critical mind, too much doubt about the documentation on the Dumbles to start a build myself. I think anybody on this forum would be dying to see anybody just do an iPhone video display of the amp they've built. Inside and out. Run over the problems they've encountered and how they solved it. The biggest plug for this forum is still Fuchs mentioning it in his interview... that's the only reason I came here. That made me believe there is info here, but it really seems to be scattered.

what's a snufa?

A SNAFU comes from military jargon, those of us that have been in or that know military types know it, I forget not everyone knows what it stands for:

Situation Normal All F*)($ed Up

SNAFU



~Phil

[martin manning]

As it appears to me, it's a negative feedback loop it sends signal directly from the anode (inverted from input) to the cathode (follows input) or vice-versa for a specific frequency due to that capacitor. anode and cathode are inverted and therefore would be negative feedback.

Yes but the cathode bypass cap dumps it all to ground, so no FB of any kind. If the cathode resistor were unbypassed there would be NFB, as Erwin mentions above.

[modman]

Hi-pass shunted to ground cuts — ie a true low-pass effectively. Now corrected more than once. Exactly.

Absolutely, when you shunt the signal path of a hi-pass filter to ground you're cutting all the highs. But this is not the case, please have another look at the schematic:

Screenshot from 2018-03-14 23-33-10.png

What is going on here is this:

Screenshot from 2018-03-17 15-14-38.png

Some of the signal is going through the 270pF capacitor, then there is a resistor to ground of 1k5, the cathode resistor.
Part of the low end goes to ground, anything under +/- 400 Hz.
I your reading, what frequencies are affected? What happens to the rest of the signal?

Rest of the signal is fed into the cathode... it's a feedback loop.

This is also a non-inverting feedback loop... it feeds signal back through the cathode. But I might be wrong...

As it appears to me, it's a negative feedback loop it sends signal directly from the anode (inverted from input) to the cathode (follows input) or vice-versa for a specific frequency due to that capacitor. anode and cathode are inverted and therefore would be negative feedback.

I agree with you Phil, it's a feedback loop, but non-inverting in my opinion signal does not invert when fed through the cathode.

The snubber cap is not in the signal path! So there's your wrong assumption.

It's not a snubber cap, but that's just terminology. Why do you think it's not in the signal path? How can any part of the signal be "shunted to ground" if the caps not in the signal path?

If you don't' believe me I would suggest you draw a schematic of a tone circuit as seen on guitar...

Are you sure?

Screenshot from 2018-03-18 19-28-27.png

In a guitar tone control, there is first a resistor. There is the resistance of the pot, and this drawing even has an extra resistor, then a capacitor to ground -- hence this is a low pass filter, when you turn it, you cut highs.

Also it's not a feedback loop. The highs are shunted to ground through the cathode bypass cap. In case of not having a cathode bypass cap and only having a cathode resistor I would agree with you.

Ok, I didn't realize that the 4.7uF was the cathode bypass cap. Don't know what the effect of that is on the 270pF feedback loop.
That 4.7uF may bleed some highs to ground too.

I still believe it's a feedback loop bleeding off anything below 400Hz and feeding the rest back through the cathode. It is a distortion circuit...

As it appears to me, it's a negative feedback loop it sends signal directly from the anode (inverted from input) to the cathode (follows input) or vice-versa for a specific frequency due to that capacitor. anode and cathode are inverted and therefore would be negative feedback.

Yes but the cathode bypass cap dumps it all to ground, so no FB of any kind. If the cathode resistor were unbypassed there would be NFB, as Erwin mentions above.

Maybe it does dump all AC to ground... but then what is the function of this part of the circuit, i.e. that 270pF capacitor?
Why is not all signal bled to ground if all AC can pass through caps?
Also, if there were no dumping to ground, it would not be a negative feedback loop, because: if you drive the cathode, the plate does not invert

Screenshot from 2018-03-18 19-55-02.png

---------------------------------------------------------
Edit 20 March
--------------------------------------------------------------

deeper theory on the 'why' and not so much on the what I can do better in this build, etc.
Please lets return to the main thread.

Here's my last answer in this thread. A lot of people had the answer: this 270pF capacitor rolls off high frequencies.

The reason I didn't accept this answer is because: nobody was able to give me a calculation of exactly which frequencies are rolled off. This is because if there is a low pass filter, there should be resistance and then cap to ground. Where is the resistance? It's in the tube and the plate resistor in parallel, assuming that's 100k // 40K makes up 28k5. That in combination with the 270pF [because 270pF in series with 5uF = 269pF] give 21k Hz as cutoff point. There is feedback to the cathode but only 1/10000 of the signal, but it is certainly positive feedback -- which is kinda crazy. If you send signal through the cathode, it does not invert to the plate. Even asked a friend who's a full time amp tech. He also knew it cut high frequency, but also thought of it as negative feedback loop and calculated it wrong.

http://www.diyaudio.com/forums/tubes-va ... whats.html

[erwin_ve]

Hi-pass shunted to ground cuts — ie a true low-pass effectively. Now corrected more than once. Exactly.

Absolutely, when you shunt the signal path of a hi-pass filter to ground you're cutting all the highs. But this is not the case, please have another look at the schematic:

Screenshot from 2018-03-14 23-33-10.png

What is going on here is this:

Screenshot from 2018-03-17 15-14-38.png

Some of the signal is going through the 270pF capacitor, then there is a resistor to ground of 1k5, the cathode resistor.
Part of the low end goes to ground, anything under +/- 400 Hz.
I your reading, what frequencies are affected? What happens to the rest of the signal?

Rest of the signal is fed into the cathode... it's a feedback loop.

This is also a non-inverting feedback loop... it feeds signal back through the cathode. But I might be wrong...

As it appears to me, it's a negative feedback loop it sends signal directly from the anode (inverted from input) to the cathode (follows input) or vice-versa for a specific frequency due to that capacitor. anode and cathode are inverted and therefore would be negative feedback.

I agree with you Phil, it's a feedback loop, but non-inverting in my opinion signal does not invert when fed through the cathode.

The snubber cap is not in the signal path! So there's your wrong assumption.

It's not a snubber cap, but that's just terminology. Why do you think it's not in the signal path? How can any part of the signal be "shunted to ground" if the caps not in the signal path?

If you don't' believe me I would suggest you draw a schematic of a tone circuit as seen on guitar...

Are you sure?
Screenshot from 2018-03-18 19-28-27.png

In a guitar tone control, there is first a resistor. There is the resistance of the pot, and this drawing even has an extra resistor, then a capacitor to ground -- hence this is a low pass filter, when you turn it, you cut highs.

Also it's not a feedback loop. The highs are shunted to ground through the cathode bypass cap. In case of not having a cathode bypass cap and only having a cathode resistor I would agree with you.

Ok, I didn't realize that the 4.7uF was the cathode bypass cap. Don't know what the effect of that is on the 270pF feedback loop.
That 4.7uF may bleed some highs to ground too.

I still believe it's a feedback loop bleeding off anything below 400Hz and feeding the rest back through the cathode. It is a distortion circuit...

As it appears to me, it's a negative feedback loop it sends signal directly from the anode (inverted from input) to the cathode (follows input) or vice-versa for a specific frequency due to that capacitor. anode and cathode are inverted and therefore would be negative feedback.

Yes but the cathode bypass cap dumps it all to ground, so no FB of any kind. If the cathode resistor were unbypassed there would be NFB, as Erwin mentions above.

Maybe it does dump all AC to ground... but then what is the function of this part of the circuit, i.e. that 270pF capacitor?
Why is not all signal bled to ground if all AC can pass through caps?
Also, if there were no dumping to ground, it would not be a negative feedback loop, because: if you drive the cathode, the plate does not invert

Screenshot from 2018-03-18 19-55-02.png

Honestly you're either a troll or plain stubborn. Sorry.

[pompeiisneaks]

I'd say, at this point, that this thread's primary purpose is to discuss this build and the discussion seems to be getting into deeper theory on the 'why' and not so much on the what I can do better in this build, etc.

Please lets return to the main thread.

I ended up destroying one of my two footswitches for the foot pedal, so I put in an order for the replacement and got the ceramics, I also noted there was another few that called out for ceramics and I've gotten those as well. I'll be too busy with work this week to do much but I'll uplaod the short video of the 'why' I destroyed it later this week, it's sad, and funny, sadly

Thanks for everyone's help and input, it's greatly appreciated.

~Phil

[dbeasley]

FWIW Phil I have great luck using the Murata Ceramic disc caps there. The 2KV rated 270pf are plentiful from mouser. And lots of different values.

Like Martin says, the cathode bypass cap negates the whole NFB loop idea of the Snubbers.

Then again, why not use them on the plates of the OD stage? DC is the same as ground to an audio signal right?

Sent from my Moto G (4) using Tapatalk

[pompeiisneaks]

FWIW Phil I have great luck using the Murata Ceramic disc caps there. The 2KV rated 270pf are plentiful from mouser. And lots of different values.

Like Martin says, the cathode bypass cap negates the whole NFB loop idea of the Snubbers.

Then again, why not use them on the plates of the OD stage? DC is the same as ground to an audio signal right?

Sent from my Moto G (4) using Tapatalk

I'd have to peek at the brand I bought, but yeah that makes sense. Once you have a cap to ground, adding another in series just makes it a different cap rating and does the needful. I get it

~Phil

[martin manning]

With no bypass cap on Rk, the snubber would actually create some positive FB, but I believe that would be overwhelmed by NFB created by Rk.

[pompeiisneaks]

With no bypass cap on Rk, the snubber would actually create some positive FB, but I believe that would be overwhelmed by NFB created by Rk.

Maybe I'm missing the obvious but anode and cathode are out of phase no? so this is negative feedback, but I'd guess almost from both sides? i.e. anode sends it's output to cathode and cathode send's it's output to anode, and they're inverted from one another? Or maybe they'd basically cancel out?

Definitely wouldn't mind understanding that interaction more.
edit: removed incorrect term 'out of phase' and replaced with 'inverted'
~Phil

[sluckey]

Maybe I'm missing the obvious but anode and cathode are out of phase no? so this is negative feedback

Yes, they are out of phase, but it does create positive feedback. Being out of phase is not necessarily the same as negative feedback. You have to consider what the out of phase signal actually does to the bias of the tube (voltage between grid and cathode). Look at the overall tube operation...

Assume the signal on the grid swings positive. This causes the tube to conduct harder and the plate voltage goes down, ie, signal swings negative. This negative signal is applied to the cathode which causes the tube to conduct even harder, which drives the plate voltage down even more (signal swings even more negative, gets bigger. Just the opposite happens when the grid signal swings negative.

Bottom line, as far as tube bias is concerned, applying a negative swinging signal to the cathode has the same effect as applying a positive swinging signal to the grid. So, if you couple some plate signal back to the grid you create negative feedback. But if you couple plate signal to the cathode you create positive feedback.

[pompeiisneaks]

I "think" I get that, I may have to reread it a few times, but that does seem to click inside the brain box

~Phil

[pompeiisneaks]

I've uploaded Video 8 in the dumble Build series. It covers the footswitch build process, as I was too busy with my day job to do much this week.




I'm also going to do another Saturday Livestream, so come check it out, 11am PST, 2pm EST. Click here to get to that stream:




~Phil