V1 Input: Grid Leak and Grid Stopper

[FourT6and2]

I'm building a Ceriatone Chupacabra. Here is its input layout. Input shield is tied back to ground at the jack.




Schematic (I think, hope it's correct):




Now, here's the schematic for how it's typically arranged (at least according to ValveWizard, et al).




So can anybody help me understand what's happening here and why Ceriatone has:

1. Placed the 27K BEFORE the 1M, not AFTER (as Aiken explains—"The resistor should be placed after the grid-to-ground resistor (usually 1 Meg or so), to avoid attenuation and to keep the signal path short. If the resistor is connected in series with the input jack and before the 1 Meg grid resistor, there is a small loss of the input signal, although, in most cases the attenuation is not enough to be concerned with (0.94 times for a 68K grid stopper and a 1 Meg grid resistor), and in amplifiers with a high and low level input, the grid stoppers also serve as attenuators.)".

Either Ceriatone is intentionally attenuating the input signal or... ?

2. What the 0.1uF cap is doing in series, after the grid stopper. Is it there as a safety, to keep voltage off the guitar? When wired up as in schematic #2, ValveWizard explains that to reduce hiss/noise caused by the grid stopper, you can lower its value and then include a small cap from grid to ground to make up for the lowered resistance. Something on the order of 100pF - 470pF. But this is not what's happening in the way Ceriatone is doing it.

What are the pros/cons of each method?

[SoulFetish]

I'm building a Ceriatone Chupacabra. Here is its input layout. Input shield is tied back to ground at the jack.




Schematic (I think, hope it's correct):




Now, here's the schematic for how it's typically done (at least according to ValveWizard, et al):




So can anybody help me understand what's happening here and why Ceriatone has:

1. Placed the 27K BEFORE the 1M, not AFTER (as Aiken explains—"The resistor should be placed after the grid-to-ground resistor (usually 1 Meg or so), to avoid attenuation and to keep the signal path short. If the resistor is connected in series with the input jack and before the 1 Meg grid resistor, there is a small loss of the input signal, although, in most cases the attenuation is not enough to be concerned with (0.94 times for a 68K grid stopper and a 1 Meg grid resistor), and in amplifiers with a high and low level input, the grid stoppers also serve as attenuators.)".

Either Ceriatone is intentionally attenuating the input signal or... ?

2. What the 0.1uF cap is doing in series, after the grid stopper. Is it there as a safety, to keep voltage off the guitar? When wired up as in schematic #2, ValveWizard explains that to reduce hiss/noise caused by the grid stopper, you can lower its value and then include a small cap from grid to ground to make up for the lowered resistance. Something on the order of 100pF - 470pF. But this is not what's happening in the way Ceriatone is doing it.

What are the pros/cons of each method?

The only way merlin would be showing a circuit in the way it was drawn in example number two, would be if he was showing an alternate way of turning the input "off". The way it is drawn there, it would shunt all frequencies to Down to -3dB at 58Hz. Of course it might function well as a triode subwoofer? My guess is it's most likely drawn incorrectly

[FourT6and2]

The only way merlin would be showing a circuit in the way it was drawn in example number two, would be if he was showing an alternate way of turning the input "off". The way it is drawn there, it would shunt all frequencies to Down to -3dB at 58Hz. Of course it might function well as a triode subwoofer? My guess is it's most likely drawn incorrectly

Maybe I drew it wrong? I mean, I didn't include the cathode bypass resistor/cap and all that. Just focusing on the grid... Pretend that 0.1uF cap is 100pF - 470pF.

Here is the image directly from his page:



Anyway, I'm more concerned with how Ceriatone is doing it and why...

[Enzo]

Having a 0.1uf to ground from the grid ought to pretty much remove all the high end from your signal, like a blown tweeter. Any grid to ground cap would be a LOT smaller.

A 0.1uf cap is large, it allows all audio frequencies through. It blocks any DC from the outside world. Most pedals have a cap on the output or otherwise do not make DC. But not all, so a big, passes everything cap lets signal through but blocks DC. Imagine if your wah pedal had 5vDC on the output. without a blocking cap, you'd have 5v on the grid, that would upset the tube bias considerably. it isn't about safety, it is about maintaining proper operation under potentially problematic conditions.

A 27k over 1 meg voltage divider would reduce the signal level by 2 or 3%, you'd never hear or miss it.

Rearranging the circuit may have potential for lower noise, but if you are not having a noise problem at the input, what is there to reduce?

Don't over analyze it, it is just a guitar amp.

[FourT6and2]

Having a 0.1uf to ground from the grid ought to pretty much remove all the high end from your signal, like a blown tweeter. Any grid to ground cap would be a LOT smaller.

Yes, please see my above post. Pretend it's 100pF or something. I just didn't change the value when I re-drew the schematic. I can redo it and upload a new photo.

My question is abut the first schematic though. And yes, the amp has a 0.1uF cap in series with the 27K grid stopper.

A 27k over 1 meg voltage divider would reduce the signal level by 2 or 3%, you'd never hear or miss it.

I'm trying to understand the how/why here. Even if it's a small attenuation, why would one choose to rearrange the grid leak and grid stopper in this way, rather than putting the 1 Meg first? I mean you're there, soldering iron in hand, it's the same amount of effort to solder the resistor here or there.

Rearranging the circuit may have potential for lower noise, but if you are not having a noise problem at the input, what is there to reduce?

I'm building someone else's design here. It's a high-gain amp. Not some Fender Champ. How do I know if I'm having a noise problem if the amp isn't built yet? RF is a real concern. I live in a big city. Lots of RF everywhere. There are radio towers a block away, and so on. But the input of this amp is wired this way in the schematic and I was hoping maybe somebody could shed some light on why someone would choose to do it this way, over having it arranged the more traditional way with the 1M-to-ground coming first.

Thanks for the bit about the cap. My question is if that cap is there to block DC from ENTERING the amp, and you say, "without the cap, you'd have DC on the grid and that would upset the tube bias considerably." But no stock Marshalls have that cap that I know of. Can't speak to modern Marshalls. But none of the ones most people think of (JMP, 1959, 2204, etc.). If it's about "maintaining proper operation," wouldn't it be present in more amps? Not trying to argue here. I'm trying to understand.

Don't over analyze it, it is just a guitar amp.

I'm building this circuit and I'd like to understand it rather than blindly following a layout...

[Chuck H]

Specific values are VERY important to circuit function. Omitting values for circuit diagrams is different. And then circuit values are typically implied and discussed. But simply installing incorrect, and grossly non functional values and expecting people to "get it" is, well, quite fetched. Please don't continue this practice. It's a waste of the incredible brain power here as interpretations are surmised and smart, helpful people's time is squandered. In fact, in that light it seems down right disrespectful. What were you thinking?

[SoulFetish]

Yes, please see my above post. Pretend it's 100pF or something. I just didn't change the value when I re-drew the schematic. I can redo it and upload a new photo.

My question is abut the first schematic though. And yes, the amp has a 0.1uF cap in series with the 27K grid stopper.



I'm trying to understand the how/why here. Even if it's a small attenuation, why would one choose to rearrange the grid leak and grid stopper in this way, rather than putting the 1 Meg first? I mean you're there, soldering iron in hand, it's the same amount of effort to solder the resistor here or there.



I'm building someone else's design here. It's a high-gain amp. Not some Fender Champ. How do I know if I'm having a noise problem if the amp isn't built yet? RF is a real concern. I live in a big city. Lots of RF everywhere. There are radio towers a block away, and so on. But the input of this amp is wired this way in the schematic and I was hoping maybe somebody could shed some light on why someone would choose to do it this way, over having it arranged the more traditional way with the 1M-to-ground coming first.

Thanks for the bit about the cap. My question is if that cap is there to block DC from ENTERING the amp, and you say, "without the cap, you'd have DC on the grid and that would upset the tube bias considerably." But no stock Marshalls have that cap that I know of. Can't speak to modern Marshalls. But none of the ones most people think of (JMP, 1959, 2204, etc.). If it's about "maintaining proper operation," wouldn't it be present in more amps? Not trying to argue here. I'm trying to understand.



I'm building this circuit and I'd like to understand it rather than blindly following a layout...

There are a few reasons why it may be desirable to use a series capacitor network on the input grid. One reason it is may be necessary is when you have to decouple the input stage is biased to a high or low DC voltage, or if the input is bootstrapped (assuming the most common method). I mean, it adds a series impedance and phase shift depending on the frequency, but I don't know why Ceriatone is doing it. It is helpful to post a schematic of the circuit you want to talk about.

[FourT6and2]

There are a few reasons why it may be desirable to use a series capacitor network on the input grid. One reason it is may be necessary is when you have to decouple the input stage is biased to a high or low DC bias, or if the input is bootstrapped (assuming the most common method). I mean, it adds a series impedance and phase shift depending on the frequency, but I don't know why Ceriatone is doing it. It is helpful to post a schematic of the circuit you want to talk about.

Here you go: http://orig05.deviantart.net/fd41/f/...lm-dbg4mbt.png

It's not just the cap I'm curious about. But why one would choose to add the cap in series as well as place the grid leak after the grid stopper, instead of before it. Thank you for any help.

[SoulFetish]

Specific values are VERY important to circuit function. Omitting values for circuit diagrams is different. And then circuit values are typically implied and discussed. But simply installing incorrect, and grossly non functional values and expecting people to "get it" is, well, quite fetched. Please don't continue this practice. It's a waste of the incredible brain power here as interpretations are surmised and smart, helpful people's time is squandered. In fact, in that light it seems down right disrespectful. What were you thinking?

Oh yeah, Chuck doesn't like it when this happens. It's a good head's up
Another helpful tip, (just throwing it out there), is never accuse Enzo of trolling..... It makes Juan angry. When Juan gets angry it frightens me.
I'm just glad we able to have this talk.

[SoulFetish]

Here you go: http://orig05.deviantart.net/fd41/f/...lm-dbg4mbt.png

Where's the power supply?

[Enzo]

You haven't built yours yet, but the Ceria people have made many of these kits. Are we aware of some problem they are all having with RF? RF is everywhere, it comes from stuff in your house, when I lived in the country I had various digital equipment in the house that threw off a lot of hash, but it rarely bothered the guitar amps.

Yes, many amps lack an input cap, Fender for example. I just explained why someone would put one there. Not everyone thinks it warrants the effort. Without that cap you would have DC on your grid - but remember that was only when some external thing had the voltage on it in the first place.

A grid stopper ought to be right at the tube pins. I don't really think of the two 68k resistors mounted to the input jacks of Fender amps as grid stoppers. That is my opinion. The two 68k resistors form an attenuator for the lower gain jack, or they even act as mixing resistors in the odd situation you plug instruments into both jacks. Using the high gain jack, the two resistors are in parallel for a 34k resistance. if Ceria uses just one jack, they may approximate the 34k with a single 27k resistor. Just my guess. It is out there as a tip of the hat to the Fender design. The 1 meg is the ground reference for the grid.

If you put the 27k right at the socket, it acts as a grid stopper and there is no signal attenuation. If you put the 27k at the jacks and the 1 meg at the tube, we get that tiny tiny bit of signal reduction. The fact is either way works, so it isn't like there is a wrong here.

I found input caps on all the Peavey schematics I opened here looking around my files. I saw none on Mesas (Mesa did use ferrite beads on the inputs though - an RF fighter), but did find them on Music Man amps. Sunn tubers lacked them. I found them on Trace Elliots. A couple Vox models and no cap. Ampeg on some and not others.

I'm building this circuit and I'd like to understand it rather than blindly following a layout...

I wouldn't want you to just blindly do anything. When I said don't OVER analyze it, I just meant that the circuitry is not precision, is not hyper critical, and will tolerate a lot of variation. Don't get caught up in any sort of it HAS TO BE a certain way, and that there is a right answer to everything that makes everything else wrong. In the glory days of Fender, every schematic had a note in the corner that all readings were within +/-20%, and in fact the resistors they used were 20% tolerance parts.

[FourT6and2]

You haven't built yours yet, but the Ceria people have made many of these kits. Are we aware of some problem they are all having with RF? RF is everywhere, it comes from stuff in your house, when I lived in the country I had various digital equipment in the house that threw off a lot of hash, but it rarely bothered the guitar amps.

Man, I really don't understand your point. The RF suppression is in the design of the amp, as Ceriatone builds it. I'm not adding anything or changing anything. I hear radio stations coming through all my amps at some point or another. Hell, I've played along with songs on the radio through my amps!

Yes, many amps lack an input cap, Fender for example. I just explained why someone would put one there. Not everyone thinks it warrants the effort. Without that cap you would have DC on your grid - but remember that was only when some external thing had the voltage on it in the first place.

Ok, understood.

A grid stopper ought to be right at the tube pins.

Yes. In this case, it is. Sort of. It's on a standoff as Ceriatone does it. But it's not on the jack.

I don't really think of the two 68k resistors mounted to the input jacks of Fender amps as grid stoppers. That is my opinion. The two 68k resistors form an attenuator for the lower gain jack,

Yep. That's what Aiken says about them. I understand that. But that's not what's happening here. The grid stopper is not being used as an attenuator for a low-gain input jack (at least it shouldn't be).

if Ceria uses just one jack, they may approximate the 34k with a single 27k resistor. Just my guess. It is out there as a tip of the hat to the Fender design. The 1 meg is the ground reference for the grid.

Hmmm, as I understand it the 27K resistor is being used—instead of 68K—to reduce noise. This is what both Aiken and Merlin discuss on their pages that I am referencing.

If you put the 27k right at the socket, it acts as a grid stopper and there is no signal attenuation.

It isn't " on the jack." But I guess it's not directly on the tube pin either. The 0.1uF cap is in series with it to the tube. And the 1M is AFTER the 27K, to ground. Instead of BEFORE it. This is the whole point of the thread. Or maybe you're saying "at the jack" to simply mean the order in which the parts appear? And not literally "on the jack itself"? I guess saying "at the jack" is the same as saying the grid stopper comes before the grid leak instead of after?

If you put the 27k at the jacks and the 1 meg at the tube, we get that tiny tiny bit of signal reduction. The fact is either way works, so it isn't like there is a wrong here.

Yes. Exactly. I understand this. That's why I started this thread, asking for help. I'm trying to understand WHY Ceriatone would choose to do it this way. That is my question here. I guess they want some attenuation??????? Because you're there with soldering iron in your hand about to connect that resistor and you have to decide to put it here or put it there. So I'm trying to understand the rationale of choosing to do in such a way that attenuates the input signal.

I have a photo of one of these amps that was modified by another famous builder. And they removed the 0.1uF cap and moved the 1M grid leak to before the 27K grid stopper. That was an intentional move. So it makes some difference?

[tubeswell]

Maybe I drew it wrong? I mean, I didn't include the cathode bypass resistor/cap and all that. Just focusing on the grid... Pretend that 0.1uF cap is 100pF - 470pF.

Here is the image directly from his page:



Anyway, I'm more concerned with how Ceriatone is doing it and why...

The context for that schematic from Merlin's page (here: The Valve Wizard ), is that he is talking about using a smaller input resistor (10k) together with a 100-470pF cap, to remove hiss while preserving bandwidth, instead of using a more 'traditional' Fender input resistor value.

[FourT6and2]

There's also this way of doing it:



This is what a friend of mine suggested:

"You can also try this as a mod later:

C10 = 22nF/630V
R25 = 1M
C17 = 100pF ceramic (against RF)
R16 = 1K

The result is way less hiss and still a good RF protection.
Also C10 can save your life if you have a faulty ground in the amp (then all B+ goes to the guitar if there isn't a capacitor between you and the amp).

C10 can be also after the 1M (soldering the 1M directly to the jack is common, so just place the 22nF directly to the jack's hot and then run the cable to the grid resistor).

This is the way today's amps are built."

[Enzo]

Yes, my shop was a few short blocks from an AM radio transmitter, I could watch their modulation envelope on my scope whenever I wanted. RF suppression in guitar amps is not meant to stop brute force RF invasions though, in my view.

I think it would help you to forget the idea of attenuation here. The 27k and 1 meg resistor make a VERY small attenuation. Like driving down the interstate at 69 instead of 70. Noise reduction or other factors are on the mind of designers, not the slight signal drop. The typical high and low jacks on an old Fender cut the signal in half, a 50% reduction from high in the low jack. That is a lot, your 2% is small.

When I talk about attenuation, I do not mean a purposeful drop in level like the hi/lo Fender system. All I mean is the the two resistors act together to lop off a tiny sliver of signal level. It isn't a useful amount or anything like a "feature", it just means there is a tiny reduction. The amp has way more than enough gain to not miss it. So when looking at your 27k/1meg resistors, know that they were not interested in obtaining any attenuation, they just didn't care that a tiny bit resulted. Whichever end of the wire between the jack and the tube you put the 27k on, any noise reduction potential it has far outweighs any 2% signal loss.

I am being literal. In a typical Fender, the jack is in front and the tubes in back with maybe 6-8 inches between. When I say at the jack, I mean like Fender did, the resistors soldered directly to the jack. When I say at the grid, I mean right onto the tube socket pin. And some might put the resistor on the part board in between.

The famous builder probably likes to build them that way. I suspect if you built it one way, then changed it to the other, you would not hear a lot of difference. Ceria has the cap in series, Merlin's does not. SO Ceria must have the 1 meg grid return on the grid side of the cap otherwise ther is no ground reference for the grid. SO assuming the cap is invisible to signal, that leaves the decision to put the 27k between the 1meg and the tube or between the 1 meg and the jack. There are good designers who would do it either way, I won't call any of them wrong. Someone may make a stronger case for one way or the other, but I think it will be a small part of the amp circuit.