Resistor Types vs Placement in Circuit

General discussion area for tube amps.

Moderators: pompeiisneaks, Colossal

FourT6and2
Posts: 178
Joined: Wed May 31, 2017 11:11 pm

Resistor Types vs Placement in Circuit

Post by FourT6and2 »

Sorry for the can of worms. And yeah, I've been reading about this sort of thing lately. But...

Carbon Composition
Carbon Film
Metal Film

Do y'all prefer certain ones in specific places in the circuit? Take a Marshall JMP/JCM800 circuit, for example. Let's say I'm using metal film in most of the spots for noise reduction and higher voltage handling. Are places where you'd still prefer to use carbon film or carbon comp? NFB resistor? Grids? Tone stack? Bias supply? Basically, I have a bunch of high-quality metal film. 1W, 1%, 500v. But not sure if I should just stick them everywhere or what. I'm not trying to emulate a vintage amp or anything.
User avatar
MakerDP
Posts: 281
Joined: Thu Jul 30, 2015 6:36 pm
Location: Central California
Contact:

Re: Resistor Types vs Placement in Circuit

Post by MakerDP »

In my builds, if it's not a 1W or higher resistor it's all metal film. Everywhere.
ChopSauce
Posts: 1038
Joined: Wed Mar 15, 2017 12:08 pm
Location: So Paris, France

Re: Resistor Types vs Placement in Circuit

Post by ChopSauce »

FourT6and2 wrote: Sat Jul 08, 2017 1:02 amI have a bunch of high-quality metal film. 1W, 1%, 500v. But not sure if I should just stick them everywhere or what. I'm not trying to emulate a vintage amp or anything.
So, you should be fine, according to AIken at least. Just one related question, if you don't mind:

- some authors recommend that the signal path be populated with low noise metal films but the Dumble way of doing things (again) is different, which seem to minimise amplification noise (through the use of low noise plate & cathode resistors- roughly) and let the noise in the signal path propagate at will (EDIT: through the use of carbon comp. resistors at the input) within those (tight) bounds - which could be summarised as : noise amplification vs. amplifiation noise (choose yours ?-)

I wish I could read any further comments about this (possibly to explain me that I didn't caught it.)
Last edited by ChopSauce on Sat Jul 08, 2017 4:22 pm, edited 2 times in total.
pdf64
Posts: 2688
Joined: Sat Mar 12, 2011 9:23 pm
Location: Staffordshire, UK

Re: Resistor Types vs Placement in Circuit

Post by pdf64 »

If you've not read RG Keen's article on the topic, then it may help to give you some rationale / direction http://www.geofex.com/Article_Folders/c ... oncomp.htm
ChopSauce
Posts: 1038
Joined: Wed Mar 15, 2017 12:08 pm
Location: So Paris, France

Re: Resistor Types vs Placement in Circuit

Post by ChopSauce »

pdf64 wrote: Sat Jul 08, 2017 1:16 pm If you've not read RG Keen's article on the topic, then it may help to give you some rationale / direction http://www.geofex.com/Article_Folders/c ... oncomp.htm
Indeed! Aiken also http://www.aikenamps.com/index.php/resi ... -it-matter speaks about the "pink-alike" noise of carbon comp. resistors (as something to be avoided.) I guess this is all about the intended use and tone of the amp. I am (tonewise) partial to amps full of carbon comp. resistors, but that's just my taste.
FourT6and2
Posts: 178
Joined: Wed May 31, 2017 11:11 pm

Re: Resistor Types vs Placement in Circuit

Post by FourT6and2 »

pdf64 wrote: Sat Jul 08, 2017 1:16 pm If you've not read RG Keen's article on the topic, then it may help to give you some rationale / direction http://www.geofex.com/Article_Folders/c ... oncomp.htm
That is EXACTLY what I've been looking for. Thank you so much for linking to that.

I'll try metal film pretty much everywhere, except the PI plates, which I'll try 1/2-watt carbon comp. And maybe the NFB resistor too.
Last edited by FourT6and2 on Sat Jul 08, 2017 5:54 pm, edited 1 time in total.
JD0x0
Posts: 550
Joined: Tue Oct 09, 2007 2:19 am

Re: Resistor Types vs Placement in Circuit

Post by JD0x0 »

I noticed something working on a Fender I'm customizing, a day or so ago.

A 1/2W MF on the plate of the input stage, followed by a 1/2W CC on the next triode will have a lower noise floor than the same circuit using a 1W CC resistor of the same value on the plate of each stage.

I tend to pretty much use MF's everywhere, except where I can use wire wound in the power section. If I want the tone of CC's on the plate, I'll still use MF on the input stages, and I'll use the lowest wattage rated CC resistor I can 'get away' with safely, to maximize resistor distortion. This is something I tried on a SF Fender I'm working on, and the noise floor is scary low. It's so quiet every time I turn the amp on, I am worried it's not putting out signal until I plug a guitar in :shock:
It's true i've lost my marbles and i cant remember where i put them
pdf64
Posts: 2688
Joined: Sat Mar 12, 2011 9:23 pm
Location: Staffordshire, UK

Re: Resistor Types vs Placement in Circuit

Post by pdf64 »

FourT6and2 wrote: Sat Jul 08, 2017 5:28 pm...I'll try metal film pretty much everywhere, except the PI plates, which I'll try 1/2-watt carbon comp. And maybe the NFB resistor too.
I think that the thrust of RG's analysis is that the best use for CCs is as plate loads in the stage or 2 preceding any power amp feedback loop.
Also I suspect that as plate load for an eg LTP, even if there's no global feedback loop, the common mode nature of even harmonic distortion may tend to cancel out and so tend to negate their contribution (hope that makes sense, I'd have a tough time trying to explain it without a lot of thinking :D ).

NFB resistors have very little Vdc across them, so using RG's rationale, there's no point using CC there.

The conditions we're looking for are -
plenty of Vdc
plenty of Vac
as late in the (unbalanced/pre-amp) signal chain as possible (for best signal to noise ratio)

Hence 2nd / 3rd stage plate loads.
FourT6and2
Posts: 178
Joined: Wed May 31, 2017 11:11 pm

Re: Resistor Types vs Placement in Circuit

Post by FourT6and2 »

pdf64 wrote: Sat Jul 08, 2017 7:41 pm
FourT6and2 wrote: Sat Jul 08, 2017 5:28 pm...I'll try metal film pretty much everywhere, except the PI plates, which I'll try 1/2-watt carbon comp. And maybe the NFB resistor too.
I think that the thrust of RG's analysis is that the best use for CCs is as plate loads in the stage or 2 preceding any power amp feedback loop.
Also I suspect that as plate load for an eg LTP, even if there's no global feedback loop, the common mode nature of even harmonic distortion may tend to cancel out and so tend to negate their contribution (hope that makes sense, I'd have a tough time trying to explain it without a lot of thinking :D ).

NFB resistors have very little Vdc across them, so using RG's rationale, there's no point using CC there.

The conditions we're looking for are -
plenty of Vdc
plenty of Vac
as late in the (unbalanced/pre-amp) signal chain as possible (for best signal to noise ratio)

Hence 2nd / 3rd stage plate loads.
I guess I read it a bit differently. The article states:

"The place to use CC's is where there's big signal - plate resistors, and ideally the stage just before the phase inverter. The phase inverter would otherwise be ideal, with plate resistors carrying the highest signal voltage in the amp, but phase inverters are often enclosed in a feedback loop. The feedback minimizes the distortion the resistor generates."

I wouldn't use CCs on the input plate stage (V1a) or the second stage in a high-gain amp (V2a) for noise reasons. The stage just before the PI in the amp I'm building is a cathode follower and just has the single 100K resistor on the tube socket as seen in typical Marshalls. The article says the PI Plates would be a fine place to use CCs. While a global NFB affects how much a CC will distort there, it still does distort a little, no?

At what point is it better to just use a MF or CF as the PI plate resistors because the global NFB loop negates any "sweetness" CCs bring to the table in that position?
JD0x0
Posts: 550
Joined: Tue Oct 09, 2007 2:19 am

Re: Resistor Types vs Placement in Circuit

Post by JD0x0 »

pdf64 wrote: Sat Jul 08, 2017 7:41 pm [quote=FourT6and2 post_id=375011 time=1499534891 user_id=17278Also I suspect that as plate load for an eg LTP, even if there's no global feedback loop, the common mode nature of even harmonic distortion may tend to cancel out and so tend to negate their contribution (hope that makes sense, I'd have a tough time trying to explain it without a lot of thinking :D ).
I think I may know of a little trick to work around the 2nd harmonic cancelation. Firstly, in Dumble type amps, as well as others, they have a trim pot to dial an imbalance to have the PI generating maximum harmonic content. Remember, if we're going for CC distortion, chances are this isn't a 'Hi Fi' amp. Perfect PI balance for harmonic and noise cancelation isn't 100% necessary for a 'good' tone.
Now to my 'trick.' use a CC on one PI plate and a MF on the other. I believe the cancelation relies on two signals of opposite phase. If only one side of the PI is generating the 2nd order harmonics from CC distortion, then there shouldn't be a corresponding negative phase to cancel it out. That's my theory, at least.
It's true i've lost my marbles and i cant remember where i put them
FourT6and2
Posts: 178
Joined: Wed May 31, 2017 11:11 pm

Re: Resistor Types vs Placement in Circuit

Post by FourT6and2 »

JD0x0 wrote: Sun Jul 09, 2017 1:42 am
pdf64 wrote: Sat Jul 08, 2017 7:41 pm [quote=FourT6and2 post_id=375011 time=1499534891 user_id=17278Also I suspect that as plate load for an eg LTP, even if there's no global feedback loop, the common mode nature of even harmonic distortion may tend to cancel out and so tend to negate their contribution (hope that makes sense, I'd have a tough time trying to explain it without a lot of thinking :D ).
I think I may know of a little trick to work around the 2nd harmonic cancelation. Firstly, in Dumble type amps, as well as others, they have a trim pot to dial an imbalance to have the PI generating maximum harmonic content. Remember, if we're going for CC distortion, chances are this isn't a 'Hi Fi' amp. Perfect PI balance for harmonic and noise cancelation isn't 100% necessary for a 'good' tone.
Now to my 'trick.' use a CC on one PI plate and a MF on the other. I believe the cancelation relies on two signals of opposite phase. If only one side of the PI is generating the 2nd order harmonics from CC distortion, then there shouldn't be a corresponding negative phase to cancel it out. That's my theory, at least.
Have you tried it?

I've definitely seen a certain someone using one, single CC resistor on the PI plate. And the other one either CF or MF. I just assumed is was a parts availability thing, though.
JD0x0
Posts: 550
Joined: Tue Oct 09, 2007 2:19 am

Re: Resistor Types vs Placement in Circuit

Post by JD0x0 »

FourT6and2 wrote: Sun Jul 09, 2017 2:40 am
JD0x0 wrote: Sun Jul 09, 2017 1:42 am
pdf64 wrote: Sat Jul 08, 2017 7:41 pm [quote=FourT6and2 post_id=375011 time=1499534891 user_id=17278Also I suspect that as plate load for an eg LTP, even if there's no global feedback loop, the common mode nature of even harmonic distortion may tend to cancel out and so tend to negate their contribution (hope that makes sense, I'd have a tough time trying to explain it without a lot of thinking :D ).
I think I may know of a little trick to work around the 2nd harmonic cancelation. Firstly, in Dumble type amps, as well as others, they have a trim pot to dial an imbalance to have the PI generating maximum harmonic content. Remember, if we're going for CC distortion, chances are this isn't a 'Hi Fi' amp. Perfect PI balance for harmonic and noise cancelation isn't 100% necessary for a 'good' tone.
Now to my 'trick.' use a CC on one PI plate and a MF on the other. I believe the cancelation relies on two signals of opposite phase. If only one side of the PI is generating the 2nd order harmonics from CC distortion, then there shouldn't be a corresponding negative phase to cancel it out. That's my theory, at least.
Have you tried it?

I've definitely seen a certain someone using one, single CC resistor on the PI plate. And the other one either CF or MF. I just assumed is was a parts availability thing, though.
Yes, I've tried it on a recent project, but did not do any comparative tests using MF/MF or CC/CC in the same circuit. It sounds good though.

I'm curious who this certain someone is, I dont think I've ever seen it before in person.. If it's one of those fellas who's amps are going for $30+K on the used market, I'd be inclined to believe the composition was selected for tonal reasons
It's true i've lost my marbles and i cant remember where i put them
FourT6and2
Posts: 178
Joined: Wed May 31, 2017 11:11 pm

Re: Resistor Types vs Placement in Circuit

Post by FourT6and2 »

JD0x0 wrote: Sun Jul 09, 2017 4:29 am
FourT6and2 wrote: Sun Jul 09, 2017 2:40 am
JD0x0 wrote: Sun Jul 09, 2017 1:42 am

I think I may know of a little trick to work around the 2nd harmonic cancelation. Firstly, in Dumble type amps, as well as others, they have a trim pot to dial an imbalance to have the PI generating maximum harmonic content. Remember, if we're going for CC distortion, chances are this isn't a 'Hi Fi' amp. Perfect PI balance for harmonic and noise cancelation isn't 100% necessary for a 'good' tone.
Now to my 'trick.' use a CC on one PI plate and a MF on the other. I believe the cancelation relies on two signals of opposite phase. If only one side of the PI is generating the 2nd order harmonics from CC distortion, then there shouldn't be a corresponding negative phase to cancel it out. That's my theory, at least.
Have you tried it?

I've definitely seen a certain someone using one, single CC resistor on the PI plate. And the other one either CF or MF. I just assumed is was a parts availability thing, though.
Yes, I've tried it on a recent project, but did not do any comparative tests using MF/MF or CC/CC in the same circuit. It sounds good though.

I'm curious who this certain someone is, I dont think I've ever seen it before in person.. If it's one of those fellas who's amps are going for $30+K on the used market, I'd be inclined to believe the composition was selected for tonal reasons
Nah, haha nothing like that.

Image
tmod
Posts: 17
Joined: Fri Mar 07, 2008 4:41 pm

Re: Resistor Types vs Placement in Circuit

Post by tmod »

No, neither Allen's or RG's amps sell for astronomical sums, but they have vast experience and are highly respected. For a reason I might add. If I wanted to duplicate a $30k amp I'd follow the maker's practices like gospel. Otherwise I'd stick with the advice of guys with proven expertise when it comes to amp knowledge. Guys like Allen, RG, Merlin Blencowe, Zoe and Iain Hartney, Celeste Hall, Alex Hall.
Gurus like Ken Fischer and Howard Dumble has some amazing amps to their names, but the science behind their work is sometimes a little, er, lacking...

Sent fra min SM-G930F via Tapatalk

ChopSauce
Posts: 1038
Joined: Wed Mar 15, 2017 12:08 pm
Location: So Paris, France

Re: Resistor Types vs Placement in Circuit

Post by ChopSauce »

I disagree. There are some rationales behind HAD's work. That is he promotes (pink) noise amplification and tend to cancel amplification noise, while the others do the opposite - in the sense that HAD' intensive use of RN65D & Mepco tube resistors is rather unique, and that Keen demonstates that most of what _should_ be the Mojo lies in amplification noise, instead.

Aiken, promoter of metal film resistors everywhere himself, wrotes that:

1. it is his personal position to manage to cancel _any_ noise ; and that

2. pink(alike) noise might indeed be pleasant to hear.

I can see no contradiction here, hence my opinion that it is just a matter of taste - if any difference could be heard.at all.
Post Reply