Resonance cap in NFB?

[daz]

I first tried this using a schematic which i have since forgotten. But i cannot recall if i put it in before or after the resistor that comes off the 16 ohm tap. So when i built this last amp i again included the cap, but i couldn't recall where it goes so i put it BEFORE the resisitor. On a side note, i use the cap without a pot. It's just full resonance all the time. So anyways, it's right after the 16 ohm tap, then the R is after that. question is, does it make any difference, and if so what? Heres the deal....i have a 100k pot and 22k R in series instead of just the 100k R itself. This allows me to vary the amount of NFB. But I'm wondering whether the cap would be better AFTER the pot/R. I could just do it, but i'd like to see if anyone can give me a reason why i should or shouldn't before i go slopping up the soldering that i so neatly did. I wanted this amp to stay clean unlike the last one where all the turrets and component leads are all sloppy looking from being resoldered a bunch of times. so i'd like to hear what ya'll think before i go doing that.

So just to clarify, the NFB loop is typicall JCM 800, but the difference is it goes 16 ohm tap>.0033 resonance cap>22k R>100k pot>on to typical NFB arrangment.

[Chuck H]

So just to clarify, the NFB loop is typicall JCM 800, but the difference is it goes 16 ohm tap>.0033 resonance cap>22k R>100k pot>on to typical NFB arrangment.

That is a very unusual feedback loop arrangement. First, unless I'm mistaken, your variable feedback circuit is creating ALOT MORE feedback than a stock 800. This is generally considered a bad thing for guitar amps. Your variable control would not allow you to even reduce feedback to the stock amount for a typical 800. OK, second, that is except for the bottom end. That resonance cap unbypassed by a pot is blocking ALL the bottom end from the feedback loop. On most Marshall type builds this would make for a very loose bottom end. So what you have is a feedback loop that is sending about 3 to 9 times the amount of NFB in the middle and top end, depending on your pot setting, and none in the bottom end. Your presece control must have a very wide swing.

I have seen many of your posts from the previous build and I think this NFB arrangement could be at least partially responsible for the trouble you were having. But if you think it sounds good now...OK

I don't think it matters if the .0033 cap is behind or in front of the resistor arrangement.

Chuck

[daz]

Chuck, i don't mean to be offensive, but everything you said is the exact opposite of how it works ! First, it DOES allow the typical marshall amount as far as the resistor because the pot is in series with the 22k. therefore i can go anywhere from 72k to 22k. True, most marshalls are 100k, but i had a 100k pot there and it gave me anywhere from 122k to 22k. But with ythe 50 i don't sacrifice anything. In fact, i ended up with a 50k because i found the last part of the travel with the 100k did nothing notable. In other words, it makes negligible difference between 72k (the total of the pot and resistor) and 100k. trust me on that. And the range of the pot is huge as far as tonal difference.

Secondly, not only does it NOT add a ton of highs and mids as you said, it does the exact opposite. Turning it down to the least resistance yields LESS highs and mids. The highs and mids bevome extremely smooth. In fact, with a normal amount of NFB the presence becomes very thin and just too much high end when cranked up as you'd expect. But as soon as you turn the variable NFB pot down the tone become awesome. the thinness goes away and compression and complexity set in and it's a fantastic sound. And lastly, the res cap does not in any setting create a lack of lows. In fact, i have bypassed it many times while testing things with a clip lead and it always loses a lot of bottom. Oh, and that loose low end you mentioned....not in the least.

So i gotta tell you, your theories about this setup may somehow be based in fact, but in practice they are so wrong it raised my eyebrows when i read your post. I mentioned it elsewhere and a few people had also done the same thing with the same great results. you may wanna try it in an amp if you don't believe me because if you don't, you're missing out ! This amp has more great tones than any marshall i've owned, and i owned a lot of mashalls since the 80-'s ! This setup does just the opposite of everything you described sonically. maybe you misunderstood me somehow, i dunno.

Oh, and the problems i have had in the past had nothing to do with this setup because i just did this within the last week. And further, it's actually what HELPED get it sounding so awesome ! It helped remove those issues even further even tho they were for the most part already gone once i got the PSU right. The amp was great before i did this, but with this it's literally floored me.

[Chuck H]

Chuck, i don't mean to be offensive, but everything you said is the exact opposite of how it works ! First, it DOES allow the typical marshall amount as far as the resistor because the pot is in series with the 22k. therefore i can go anywhere from 72k to 22k. True, most marshalls are 100k, but i had a 100k pot there and it gave me anywhere from 122k to 22k. But with ythe 50 i don't sacrifice anything. In fact, i ended up with a 50k because i found the last part of the travel with the 100k did nothing notable. In other words, it makes negligible difference between 72k (the total of the pot and resistor) and 100k. trust me on that. And the range of the pot is huge as far as tonal difference.

I'm going to try not to be insulting...(or insulted)

YOUR feedback adjustment INCREASES negative feedback. It does not decrease it. Decreasing feedback is the usual goal for such controls. At the 22k adjustment you are using a 4:1 feedback ratio. A typical 800 is 20:1. More negative feedback means less gain and harmonics.

And it gets worse...Your using the 16 ohm tap. A stock 800 uses the 4 ohm tap. Since the 16 ohm tap is good for roughly twice the voltage of the 4 ohm tap you would need to use roughly 200k in the loop to have the same amount of NFB as an 800. But you actually decreased resistance instead of increasing it.

YOU ARE USING ALOT MORE NRGATIVE FEEDBACK THAN YOU THINK YOU ARE. And way too much. Especially in the mids, where most players want more harmonics for good tone.

Secondly, not only does it NOT add a ton of highs and mids as you said, it does the exact opposite. Turning it down to the least resistance yields LESS highs and mids. The highs and mids bevome extremely smooth. In fact, with a normal amount of NFB the presence becomes very thin and just too much high end when cranked up as you'd expect. But as soon as you turn the variable NFB pot down the tone become awesome.

OK. I never said your tone would have a ton of mids and highs. I said that you were feeding back mids and highs. That would result in the opposite effect. As you've noted.

And lastly, the res cap does not in any setting create a lack of lows. In fact, i have bypassed it many times while testing things with a clip lead and it always loses a lot of bottom. Oh, and that loose low end you mentioned....not in the least.

Again, I didn't say that. I said you weren't running any low end through the feedback loop. And again, that would have the opposite effect. As you've noted.

It seems to me that your doing an awful lot of jimmying the knobs to find a psuedo normal sounding spot.

So i gotta tell you, your theories about this setup may somehow be based in fact, but in practice they are so wrong it raised my eyebrows when i read your post. I mentioned it elsewhere and a few people had also done the same thing with the same great results. you may wanna try it in an amp if you don't believe me because if you don't, you're missing out ! This amp has more great tones than any marshall i've owned, and i owned a lot of mashalls since the 80-'s ! This setup does just the opposite of everything you described sonically. maybe you misunderstood me somehow, i dunno.

Oh, and the problems i have had in the past had nothing to do with this setup because i just did this within the last week. And further, it's actually what HELPED get it sounding so awesome ! It helped remove those issues even further even tho they were for the most part already gone once i got the PSU right. The amp was great before i did this, but with this it's literally floored me.

I'm sorry, but I think it's you that misunderstood me (and misquoted me). I was trying to help. In fact, I'm the guy that put you onto the Hammond transformers in a previous post.

Let me say this. If you like the way your amp sounds, then it's right for you. But please try this: Put the feedback loop on the 4 ohm tap. Temporarily bypass the .0033 cap with a small value resistor (1k or less). Turn your feedback adjustment up full (72k). Run the presence control at a lower number than you have been and crank up your amp. That should make a believer out of you. If you like it you can rebuild a more typical circuit. If not then it's easy to put it back the way it was.

Cheers

Chuck

[Enzo]

Is everyone talking about the same thing?

Unless I am missing something, doesn't the amount of feedback increase with a lower feedback resistance? The more feedback, the lower the gain of the power stage? freq selective considering the caps of course.

In any case, I chimed in to say that when a resistor and a capacitor are in series, and nothing connects to the middle between them, then it doesn't matter which order they are in. If you order a burger and fries you get the same thing as when you order fries and a burger.

[daz]

And it gets worse...Your using the 16 ohm tap. A stock 800 uses the 4 ohm tap. Since the 16 ohm tap is good for roughly twice the voltage of the 4 ohm tap you would need to use roughly 200k in the loop to have the same amount of NFB as an 800. But you actually decreased resistance instead of increasing it.

there are marshalls that use both. I know because i looked over many schematics when adding the NFB

More negative feedback means less gain and harmonics.

Not LESS harmonics...more ! Yes, less if you turn it down and nothing else. But i compensate for it by adding more hi end, and the net result to my ear is a even richer tone and more compressed. less gain, sure. But who cares...the amp is loud as heck so if anything thats a bonus.

YOU ARE USING ALOT MORE NRGATIVE FEEDBACK THAN YOU THINK YOU ARE. And way too much. Especially in the mids, where most players want more harmonics for good tone.

Look, this kind of statement assumes i'm getting crappy tone and my ears aren't good enough to hear it. Sorry, but thats just completely wrong. I gigged constantly w/o stopping except for a short period from 1980 too the mid 2000's and owned uncountable amounts of gear during that time including a truckload of amps. i was anal about tone and always got what others felt was great tone. I may be no tech, but i know tone and i know what OTHERS like too. To assume i'm getting bad tone the way i'm using it you have to be assuming my ears are bad. If thats the case we may as well end this right here. i don't know why your theories don't jibe with my results, but they certainly don't. Others have done this and find it as in one guy's words, "very useful".

It seems to me that your doing an awful lot of jimmying the knobs to find a psuedo normal sounding spot

See, here again you simply don't even listen to what i said or don't believe me. Read this: I DO NOT HAVE TO "JIMMY" WITH IT TO GET THE SOUND RIGHT !!!! Let me try and explain this again....the amp has more good sounds than any marshall i've ever owned. The NFB setup allows me to dial in MANY GOOD TONES, whereas other marshalls to my ear have always had ONE setting where i liked them and deviating from that would yeild tones i felt were not usefull. This one has 3 or 4 totally different and all equally delicious tones and everything inbetween. Let me make this clear and hopefully this time you'll believe me....the amp no longer had even a trace of any of the issues i had before. NOT A TRACE. Please, if you are going to reply again and you retain little of what i say, at least retain that because if you don't then you're talking about an amp that no longer exists. So no, i absofreakinglutly do NOT jimmy with the knobs to find a good tone. Now i play with them just to dhange sounds because there are several really beautiful ones i can get between the NFB and it's adjustments and no NFB.

But please try this: Put the feedback loop on the 4 ohm tap. Temporarily bypass the .0033 cap with a small value resistor (1k or less). Turn your feedback adjustment up full (72k). Run the presence control at a lower number than you have been and crank up your amp. That should make a believer out of you.

Hmmmm...again, it's been like that. It was like that forever ! And again, i did this just last week and it improved the amp's versatility immensely. But I realize i can tell you that till the cows come home and you aren't going to listen.


finally, i don't really care what the technical reasons are unless i'm going to blow the amp up. If you can tell me that may happen, then i'll post the question here and see what others say. But till you give me reason to believe that, how can you argue with fantastic results? Aside from the amp blowing up, assuming you feel thats gonna happen, the only other reason for you to argue is that you believe i can't hear. And again, if thats the case lets just leave it here because that will really start a flame war. I know tone, period. I'm not going to argue with you on that because you have zero knowledge of what i hear. I have 100% knowledge of it. To sum up, you have to assume i know tone, and then you have to give me a good reason why i shouldn't do this. I wouldn't tell you to stop eating something because I don't like the taste. But i would if i knew it was poison.

[Chuck H]

Look, Daz

I've seen this pattern with you before. Advice is sometimes contrary to what you want to hear. But instead of taking it into consideration, you become defensive and even insulting...And at least a little wrong. You strike me as the kind of person that is a poor loser at Monopoly, right? How'd I do?

If you like it then it's right for you. I offered an opinion based on whats typical in circuitry like this to get what most consider to be good tone. Your opinion can differ. Thats OK. It doesn't make you wrong. I'm not saying your wrong. I'm just saying that what your doing is contrary to what has worked for most. The difference is subtle but significant. And I'm not going to speculate as to why you have trouble with the distinction.

Cheers

Chuck

[Chuck H]

Is everyone talking about the same thing?

Unless I am missing something, doesn't the amount of feedback increase with a lower feedback resistance? The more feedback, the lower the gain of the power stage? freq selective considering the caps of course.

In any case, I chimed in to say that when a resistor and a capacitor are in series, and nothing connects to the middle between them, then it doesn't matter which order they are in. If you order a burger and fries you get the same thing as when you order fries and a burger.

Enzo...Boy I feel kind of silly having gotten involved in this one. I wouldn't ask you to re read it, but I'm pretty sure I stayed on track. I hope it doesn't look as if I'm the confused one.

I did mention that he could use the cap in front of or behind the resistors in my first response.

Chuck

[daz]

Ok, you seem to have had your final say, so heres mine. You're a tech, i'm not. But the problem i was having with you is that you are doing something i've seen before and really is irritating. More on that in a minute. Speaking of being a bad loser, heres what i see.....you refuse to admit 2 very important things....1)-i have not had those issues you mentioned since i got the voltages right in the pre. 2)-you assume the amp sounds lousy and i'm somehow fining a spot by jimmying the knobs thats not too bad, but that i will soon no longer feel it sounds right.

You're wrong on both counts, but you have no argument unless those things are true. And thats what is irritating to me. The fact that you are trying to justify your theories by assuming that those problems still exist and/or that my hearing is either fried or so off from everyone else's that the amp actually sounds like garbage. So in other words, you don't believe me. I told you already that i'm not questioning your theory, i just said that "i don't know why it doesn't jibe with my results". "I don't know why" is the key there....i did NOT say your theory is full of s**t, i said i don't know why it's different from my results. If i thought your theory was BS i would KNOW WHY if you get my drift. So the point is, you have to assume i'm lying to say what you did. I only have to question theory verses results which anyone here will tell you do not always make sense. So you think i'm a lier in order to prove your theory, i think your theory simply doesn't reflect my results. Whos the bad loser here? Don't forget....you can assume as much as you wish by judging me on things i've posted in the past, but i'm the one sitting here with the amp listening to it, not you. Try not to assume that things never change. Things DO change, and things eventually are finalized. This amp is done, end of story. Will i ever F with it again? who knows, i enjoy tinkering and i might. But it won't be to fix flaws in it. I tinkered with the best amps i ever owned. the point is whether you choose to believe it or not, the amp is now without flaws, period. And i am finally working on cosmetics with no desire to touch the circuit.

Finally, as to your boy who cried wolf theory, remember this.....the TRUTH was that there finally really was a wolf, but no one believed him. Those who didn't believe him weren't there to see the wolf, so thier assumptions were wrong. Doesn't matter whether they felt they had reason to disbelieve it, they were wrong because they weren't there and they did not know the truth. As much as you think you know me from my postings, you'd be a fool to believe you do because you cannot possibly get a feel for who someone really is by reading what they say in type.Thats the flaw with forums and why there is so much flaming.

[Enzo]

I think the key to effective posting is to ignore any perceived inferences and focus on the circuits.

[daz]

Well, you may be right. But when someone tells me my amp has to sound like junk according to his theory when i'm the one sitting here listening to it and i'm in absolute tonal heaven, it's pretty tough to ignore that ! i mean, you know what i went thru with this thing. After all that i'm finally truly and completely floored by this thing and someone tells me it cannot possibly sound good. Kinda irritating don't ya think? I put several months of my time sweat into this thing and the finalization that i felt would never come turned out to be better than i had hoped, then this. Your pretty immune to that kind of thing from what i can tell Enzo, but i'm not. Wish i was, but i'm not.

[Alex/Tubewonder]

IMO this discussion leads to nowhere. Based on Ops verbal description I can draw three different configurations of his NFB circuit. So, what circuit is being discussed?
Just pointing out that it is possible to visualize different things from a verbal description of a circuit.
Could the OP post a schematic, please?

[daz]

I don't have a schematic, but it should be very simple to visualize. OT 16 ohm tap>.0033 cap>22k R>50k pot. From that point is goes to a typical marshall JCM 800 NFB loop. In other words, all those components listed are in series right where the NFB resistor would normally be, coming right off the OT tap.

[Chuck H]

I think this is it. I included a typical JCM800 loop. Note the OT tap. The 800 uses the 4 ohm tap and Daz used the 16 ohm tap. The example is a two power tube amp. But Daz may have a four power tube amp. So it's worth noting that the Marshall 1959 100 watt uses the 16 ohm tap (though a popular mod is to swap it to the 4 or 8 ohm tap). Daz, let me know if this is wrong.

Chuck

[daz]

Almost like the first one except the signal from the OT, the cap at the unused side of the PI, and the PI tail are all connected to a 4.7k, not ground. the 4.7k is then grounded. so those components are all separated from ground by that resistor like in a 2204.