Seem to have found something interesting, need explanation

[daz]

As some of you may know from my past posts, a couple things i've long been trying to perfect in this amp is getting my preamp gain a bit lower w/o losing the current harmonic complexity it has, and getting max dynamics. I don't want to get excited prematurely, (yeah, i know the jokes welling up in your evil brains) but i seem to have maybe found one simple tweak that accomplished both. I always thought the extra gain i could never seem to get rid of w/o killing the tone was in the preamp, but it seems to be the PI ! What I don't understand is exactly what about the PI's circuit is the same as a gain stage. But i know if you change the bias on a gain stage by raising the cathode R value gain goes down, at least if it's not bypassed. I figured the PI should be the same, and so i added some resistance to the 470R cathode. This is like the typical 2204 PI, 100k/82k, 1M/1M. 470R/10k.

So now i have just about 2k instead of 470R and i swear the dynamics are off the hook. If i turn the guitar down to 5 it cleans up as much as it did before at 3. Picking lightly then hard gives you a clean to dirty range that is probably double as wide as it was. Maybe i'm having one of those ear moments, but it was rather surprising. I have to wait till tomorrow to really give it a long workout. But can someone explain this to me? I understand what it's doing as far as the bias change and lower gain. But why am i getting a much wider clean to dirty dynamics and less OD at the PI which shouldn't even be distorting at bedroom volumes with a pre PI master? The OD i get with my guitar on 10 is probably 20% lower. But turning it down really drops the OD and cleans up way way more than before even if i turn the gain up to have the same amount i usually played with b4. I don't get how thats happening at the PI at low volume playing. Am i spacing or is there a reason for this?

[capehead]

This is exactly what I was saying in my post about the PI distorting. The Marshall PI is very dirty, which sounds great when you don't have much preamp distortion to muddle up the harmonics it creates when clips, not so great if you want to hear more preamp distortion. Your solution is somewhat intriguing though, but the downside will be less overall output when you want to crank it up. That is why I am looking into constant current PI's since it balances the PI better, similar to what you did by increasing the 470R to 2k, except it doesn't sacrifice gain in the process.

[daz]

This is exactly what I was saying in my post about the PI distorting. The Marshall PI is very dirty, which sounds great when you don't have much preamp distortion to muddle up the harmonics it creates when clips, not so great if you want to hear more preamp distortion. Your solution is somewhat intriguing though, but the downside will be less overall output when you want to crank it up. That is why I am looking into constant current PI's since it balances the PI better, similar to what you did by increasing the 470R to 2k, except it doesn't sacrifice gain in the process.

Really?! I figured the PI was clean as a whistle if you have a pre PI master set low. Not sure why it's so distorted with a low signal going into it, but thats something i wish i knew earlier. If no one mentions something like this it doesn't occur to me because it's not obvious at all. As for the output being less.....bonus #2 ! You kiddin....my amp is so friggin loud i can't get it past 8:00 at home or i'm at stage volume. So this is yet another improvement. I have tweaked this amp endlessly for years and a large part of what i was trying to accomplish is right there in the gain of the PI ! It seems to even sound richer and more balanced. Plus it seems to have changed the NFB tone and made it more open sounding and useful. I bet it even fixes the issue i mentioned in another thread where the NFB tends to go away the more i turn up. In short, it may have fixed several issues that have been dogging this amp all along. I can't wait to give it a real workout tomorrow. I only hope it's not one of those tricks my ear is playing on me.

[Chuck H]

Daz,

I can only see your result happening if either the power tubes are being clipped by the PI or the PI itself is clipping. Are you getting these results with your master volume down or at higher volume?

The PI can be a very sensitive thing to adjust. The bias point is pretty important for getting accurate signal that will drive the power tubes to full power. Biasing cold, as you have, could mean that you're either hearing the cold clipped ends of the PI wave form, which would normally not be heard because the power tube should be in cutoff at that time, or you've increased the headroom on the positive swing enough that the PI's dynamics on the positive swing are no longer clipping. Most likely some of both is happening. It may sound fine the way you're using it but it may not work as well for other operating conditions as it did. There may be other adjustments to the PI circuit that can be made to get you into a similar tone that wouldn't affect the overall performance of the amp so much. But I'll need to know the testing conditions to get a better picture of the problem we're solving for.

[Chuck H]

Daz,

If you're pre PI master is set low your PI is clean. Ignore that notion before it clouds things further.

As to the apparent loss of NFB with volume, all I can do is speculate that it's a misperception and the differences you're hearing are due to other parameters. The NFB remains analogous to the amps total gain right up to point of PI and power tube clipping. So try to ignore that notion too. It's clouded things up enough already.

[capehead]

Daz,

I can only see your result happening if either the power tubes are being clipped by the PI or the PI itself is clipping. Are you getting these results with your master volume down or at higher volume?

The PI can be a very sensitive thing to adjust. The bias point is pretty important for getting accurate signal that will drive the power tubes to full power. Biasing cold, as you have, could mean that you're either hearing the cold clipped ends of the PI wave form, which would normally not be heard because the power tube should be in cutoff at that time, or you've increased the headroom on the positive swing enough that the PI's dynamics on the positive swing are no longer clipping. Most likely some of both is happening. It may sound fine the way you're using it but it may not work as well for other operating conditions as it did. Thee may be other adjustments to the PI circuit that can be made to get you into a similar tone that wouldn't affect the overall performance of the amp so much. But I'll need to know the testing conditions to get a better picture of the problem we're solving for.

In your opinion, what bias point is optimal in both linearity and output?

[Chuck H]

In your opinion, what bias point is optimal in both linearity and output?

Whatever provides the power tubes with the necessary drive all the way up to clipping and also has the power tubes in cutoff before the PI clips on that cycle. It can be different for different power amps.

[daz]

You may be right Chuck, but last nite i was pretty awe struck by this. But it wouldn't be the first time i plugged in the next day and said "what happened?". Hard to imagine that after the results i heard, but i will reserve my thoughts on what i heard being real till i get some playing time in tonite.

I was of the same mind as you on this, that the PI should be clean unless the signal going in is slamming it. And i couldn't imagine that being the case as such low volumes. But i couldn't ignore what i was hearing, and it spoke to many things i have always felt wasn't right with the amp. The main and biggest question i have had about it from day one is this....it felt like no matter how low i set the gain knob or how low the guitar volume, the tone would never seem to get clean. I could turn the gain down and it DID get cleaner as i turn down, but even way down at say 9:00, if i hit the strings hard it would distort. It just always felt wrong like that. And no matter what i did to the preamp, and you know how many configurations I've tried chuck, it always seemed too dirty. Even the clean sounds seemed like there was something dirty about them even if not actual OD. Hard to explain. But it always seemed there was some source of distortion, BAD distortion elsewhere. And heres another thing that seems to corroborate whats going one....when i tried post pi masters it would always sound utterly horrible with ratty distortion that you'd consider unusable once turned up a bit.

Not to the point you make that theres better ways to do this. You may be right, but one thing makes me question that. If u recall i recently installed a bootstrap master and made a thread about it. But i finally dispensed with that and went back. But while i had it installed i did NOT get this cleaner tone that is dynamic and just right like i think i heard last night. If it were just a matter of lowering gain in the PI that should have done the same thing right? So maybe there is something else going on by lowering the gain in the PI this way thats more important than just the loss of some gain. But again, tonite i will be able to see if what i really heard is real or me having a twilight zone moment.

EDIT:I'll try and upload a schematic in a bit once i revise several things on it.

[daz]

Here's a schematic i just edited. I may have missed some revisions because after i uploaded it i already noticed one. That being there should be a on/off switch on the presence. I think otherwise it should be right.

[teemuk]

Daz, have you reconsidered my advise of acquiring that oscilloscope? It would once again prove its worth in so many ways.

[g1]

Couldn't agree more with Teemu. You are light years beyond being able to get away with just a meter just for biasing or some such.
Having more firm data would show you whether your ears are hearing the same today as yesterday etc.
Being equipment limited, you not only do yourself a disservice, but the rest of us too .
The kind of insights and ideas you have need to be archived as real hard data.

[Chuck H]

Scopes for us are cheap. We're not working in serious minutia and we don't need ultra HF capabilities. You can usually find a good, working used scope good to 20MHz with probes on eBay for less than $200!!! Mine even came with the owners manual I won't say I use it as effectively as I could, but it's pretty damned easy to be more effective than I was without it. With it you'll need something to generate a test signal and also a dummy load. Another $150 maybe. For $350 you can take it to the next level. daz would enjoy it immensely.

[daz]

I was going to a few months back and even mentioned it here. Theres a cheap (i think it was $50)old goldstar i posted about at this surplus place. But when i went to possibly pick it up it only had one probe and i wasn't sure if that was a problem or if it's supposed to be like that. And the probes on scopes i've seen are not just conductive simple conductors are they? Or is that all you need and the unit just has to be grounded to the work?

[Enzo]

I use dumb old cheap scope probes, they work fine for me. $30-40. If I buy a used scope, I expect to have to buy probes for it. Even if it comes with probes, they are usually well used and tired.

The thing is, until you start using a scope, you don;t even know what you are missing. Imagine you found a forum like this one, but it was in Russian or CHinese. You could learn a lot looking at the images, get some schematics, see the occasional formula, but until you learned to READ that language, you;d have no idea of the wealth of knowledge being shared. A scope will open your eyes and make you aware of much that has been beyond you so far.

We become what we behold. We shape our tools and then our tools shape us.

Marshall McLuhan

[daz]

I use dumb old cheap scope probes, they work fine for me. $30-40. If I buy a used scope, I expect to have to buy probes for it. Even if it comes with probes, they are usually well used and tired.

The thing is, until you start using a scope, you don;t even know what you are missing. Imagine you found a forum like this one, but it was in Russian or CHinese. You could learn a lot looking at the images, get some schematics, see the occasional formula, but until you learned to READ that language, you;d have no idea of the wealth of knowledge being shared. A scope will open your eyes and make you aware of much that has been beyond you so far.

Ok, but what do i need besides that one probe? Do i need 2? Are they teh same? I know ZERO about scopes aside from the fact they show wave forms of a signal. I don't know what the probes must be, but i did notice that one had a 3 way switch on it. Don't know if i'd need another one of those or jusat a simple probe like a meter or what.