cathode biased + NFB?

[daz]

Quick question....is there some general consensus that using NFB with a cathode biased amp is not good, and if so why? Seems i have seen this said several times. This as some of you know is how my EL34 build is and it seems to sound better with NFB engaged. when i switch it out the tone becomes more harmonically rich, but the low end is slop. So i'm probably going to trash the switching and just go with NFB permanently, but i'd like to hear what you all have to say about this first.

[Jazz P Bass]

Cathode bias NFB

A lot of amp manufacturers use NFB on a cathode biased output section.
If it sounds better, then it is doing its job.
Don't get caught up in the" NFB is bad" routine.
Used properly it's great.

[daz]

Oh, trust me....i have no seconds thoughts about that at all. I love NFB. It creates a tightness that for me is mandatory. I was just wondering why it is i have read several times that NFB shouldn't be used with a cathode biased output. But i had the amp fixed bias before but thought it sounds better CB.

[Chuck H]

I think it's just lore, mojo and snake oil. One of my favorite designs uses cathode bias and NFB. It's just one of those things because there are a couple of notoriously good sounding vintage amps that use cathode bias and no NFB so the lore follows that with cathode bias you shouldn't use NFB. This logic (?) can't apply to any design other than those particular vintage circuits. Anyone who thinks this observation makes for a universal truth is short sighted and/or not very experienced with actual circuits.

For what it's worth there is a similar sect that believes that "cathode bias" = "class A". I've even seen amp techs fail to make this distinction even though they are two unrelated specifics.

Chuck

[Steve Conner]

The more RC time constants (aka "poles") you have inside a feedback loop, the harder it is for the designer to get it stable.

Two is easy, but with three, motorboating beckons: consider the standard tremolo oscillator circuit, three RC time constants inside a feedback loop.

Four or more, and your slide rule and copy of the Radio Designer's Handbook are in for a workout. It can be done, the Williamson amp has four, but it involves a fair bit of maths.

There are already three poles in a standard power amp: PI input coupling cap, power tube grid coupling caps (the two count as one pole) and output transformer magnetizing inductance. (RL time constants count the same as RC.)

And cathode bias adds a fourth one. I seem to remember Daz had trouble with motorboating caused by this in the past.

That's my $.02 on the problem anyway.

[Tubenit]

cathode biased and NFB

I use a spdt switch and NFB on all my cathode biased amps. Works fine with no issues.

With respect, Tubenit

[daz]

I use a spdt switch and NFB on all my cathode biased amps. Works fine with no issues.

With respect, Tubenit

On a unrelated note, what devices do you use in that effects look? i tried a passive one there too, but it wasn't very usable with much of anything w/o one issue or another. Any tips on how to make work that aside from what i'm seeing in the schematic?

[Tubenit]

passive effects loop

The only thing I ever use in the passive effects loop is a Boss DD3 digital delay which sounds great there. Very spacious and noise free. I've tried other effects like tremolo or overdrive/distortion which don't sound right to me in a passive loop. I have the same passive effects loop on all 3 of my amps just for the DD3.

With respect, Tubenit

[JoeM]

tubenit,

In the schematic you posted, V1 is a pentode, but the reference chart for V1 shows it as a 5751. Am I missing something?

[edit] From the schems title, I think you meant that to be a 5879.

[daz]

The only thing I ever use in the passive effects loop is a Boss DD3 digital delay which sounds great there. Very spacious and noise free. I've tried other effects like tremolo or overdrive/distortion which don't sound right to me in a passive loop. I have the same passive effects loop on all 3 of my amps just for the DD3.

With respect, Tubenit

hmmm....thats all i use for effects aside from occasional chorus. I hsve a 1/2 rack size DSP that didn't work well there. i use a second amp as effect with it, but i would like to be able to use my main amp only with a digi delay in a loop like that for less important things like smaller gigs/parties/etc. I may have to think about that. Does it seem to be in the least harmful to the amp's tone?

[Tubenit]

effects loop

Yes, the schematic showing 5879 in V1 is correct (it's not a 5751).

I can not tell that the DD3 in the effects loop colors the tone at all. It seems to just add delay to it.

The DD3 going into the input jack was intolerably noisey to the point of being unuseable for my purposes. In the effects loop it's incredibly quiet.

With respect, Tubenit