Tweet
I have a .68uf bypass on the first 2 stages of my 3 stage cascaded preamp and i'm often told i have too much gain. And i agree and have long tried many ways including split loads but none have left the tone exactly the same. Removing one of the .68's to low gain always changes the tone radically away from what i want and If i remove it and try and add highs elsewhere still no joy. Plus then the gain goes TOO low. So in the end what i always wished for was less gain but w/o killing the high end those .68's give me. So what i did was put the .68 on V1B in series with a resistor, trying a few till i got the amount of gain i wanted. I noticed the high end didn't even seem to be affected until i turn the guitar down a bit and then it actually attenuates the top just right, so it actually helps the EQ there. When the guitar is all the way up i don't seem to notice a loss of highs at all. I guess i just posted this because i'm curious whether anyone has seen this done before. If i have seen a schematic with it i don't recall. Is there some downside to thins or do some other amps implement this same design ? It sure seems to be the perfect gain reduction for me where everything else imaginable I have tried has had side effects.
-
-
Maybe your problem isn;t too much gain, it is too much signal? With all these stages, did you include any voltage division to reduce signal level between stages?Education is what you're left with after you have forgotten what you have learned. -
Yes, thats what i meant but i guess i just don't understand the lingo or the difference. I always think of gain as the strength of the signal. And i HAVE had voltage dividers, but at the moment i only have a gain pot and at the 2>3 area just a load as recommended here. Neither way got me what i wanted exactly because it's not that i couldn't reduce gain enough, it's that those methods always changed things too much.(and yes, i used treble bypass on the dividers when needed) This does it w/o side effects from what i can tell so far.Originally posted by Enzo View PostComment
-
Putting a resistor in series with a cathode cap is perfectly normal and is a recommended gain reduction technique in a number of published resources. I don't recommend it, but I've even seen folks put a pot in series with cathode caps so that the gain can be adjusted from the front panel. I forget which model it is, but I recall seeing what I think was a Marshall(?) schematic using a .1 cathode cap going to a pot with a grounded wiper with the other end of the pot tied to a cap in the signal path. It was used as the treble control. Fully clockwise the cathode bypass cap was grounded (maximum treble gain) and at the other end that cap was effectively bypassed and another cap (can't recall value) is shunted to ground like a typical treble bleed cap. Pretty slick, unique control. The problem with putting any pot on a cathode bypass cap is usually noise. It is really hard to run wires to and from the front panel of an amp without introducing noise and making a hi gain stage get unstable and want to oscillate and squeal. It certainly can be done, but you just have to be very careful about exactly how it is implemented to make it work in your circuit.
To reduce gain I am sure you have tried all of the "normal" things most folks suggest, but how many less common things have you tried? One thing that comes to mind is actually using *shudder* negative feedback around one or two stages. You'll see it in some Dumble-type amp schematics. Try using a 10M-30M feedback resistor from the plate back to grid. You will have to block the DC by either picking the sognal off after the plate coupling cap or use a completely different cap just for the feedback. In the latter scenario you get the additional ability to make the feedback frequency dependent; maybe you only need to add feedback to the upper register of the signal and not the whole thing? Experiment and see. Alternatively, you could also put the feedback resistor/cap directly between the plates of two successive stages or from grid to grid of two successive stages. The principle is still the same, but I'm sure each will affect the tone in a slightly different way. Maybe one of them isn't so bad.
On the stage with the partially bypassed cathode resistor you could send feedback from the plate of the following stage back into that cathode. Remember, signals amplified using the cathode as the input get amplified non-inverted through the stage. This puts the feedback "outside" of the direct signal path (sort of). It wouldn't work well on a fully bypassed stage since the cathode cap would just shunt most of the signal to gorund. With this method you would have to use a much smaller feedback resistor (better for noise anyway) because it will form a voltage divider with the cathode resistor (hmmm, sounds an awful lot like our normal presence controls in the output stage, don't it?).
I assume you've also tried a split plate load resistor? Instead of taking the signal right off of the junction of the plate load resistor and plate, split your 100K (or whatever) plate resistor into 90K + 10k or 80k + 20k or 50k+50k, etc. and take the signal off of the junction between the two resistors. It just turns your plate resistor into a voltage divider. Theoretically, this should have the least effect on messing with your tone. I have never had great luck with it though. Maybe just shaving a tiny bit of signal off of each stage with this technique would get the right results?Comment
-
Yep, done all those, some many times especially the split load. To tell you the truth, i went back to the previous config because it seemed like it WAS affecting some things in what for me was a negative way. I think the change in the gain post as to where i had to set it for the same amount of gain i use normally changed things due to the pot's inherent way it changes the tone at different levels, especially considering it has a bleed cap. So if i turned it up a bit to make up for the lost gain some of the smoothness and feel changed. The cleaner tones when turning the guitar down were maybe better tho. But overall not. I think maybe the amount of OD is just right the way it way. My real reason for doing this in the first place was in hopes of making the gain pot to work in a more natural way. It doesn't get very clean when way down. Even with the pot way down, you hit the strings hard and it still distorts more than i think it should compared to most amps like this. I always figured that was due to too much gain....but there we go again....too much signal or gain? I always look at it as signal strength and don't understand the difference even tho i know it's very simple theory. I suppose the answer lies in there somehow.Comment
-
Have you tried dual gain pots between stages 1&2 as well as between 2&3? Individual pots are fun because you can really dial in just where the distortion sets in and cleans up. A single dual gang pot would still help your issue of getting it to clean up. Between Stages 2 & 3 you could even get away with just using the pot as a variable resistor in parallel with the grid leak resistor so that you don'e introduce and additional series resistance. When fully up you would have your full, unaltered distortion but as it turned down you would be increasing the effect of the voltage divider, thus reducing the signal to stage 3 and helping it clean up faster.
Been there, done that too I suspect!
What did you find with the feedbacK? Too "un-tubey"?Comment
-
Yeah, in fact the reason theres no gain pot there is i found it useless. I'd set it somewhere and it never sounded as good at any other position. So i removed it and tried it a few times since with the same results. In fact thats how it was when i 1st built it. Seems like the more open i leave the signal there the better it is dynamically as far as turning the guitar down to clean up. So i just have a 470k load at that point now instead of a divider, and a 100k grid stopper. But regardless of whether theres a voltage divider there or just a load, turning the gain pot on the earlier stage down doesn't seem to make low settings get as clean as i'd expect, especially given that this is not a real high gain preamp at all. maybe about as much as a peavey classic 30 if thats something u r familiar with. Or maybe 20% more drive than a blues Jr. But talking to you about this has given me an idea i haven't yet tried.....maybe remove the gain pot and put a load there and put the gain pot where the load now is. In other words, swap them. Load at 1>2 and gain pot at 2>3. Hmmmm...i think i will try that.Originally posted by cbarrow7625 View Post
Not sure what u r asking RE: "the feedback"?Comment
-
Stranger designs have been successfully commercialized! Not sure why you would go back to a pot between stages 2&3 if you already determined it didn't really do what you wanted. Sometimes "simplifying" the design and leaving yourself with less "real time tweaking" options can actually be helpful in situations like this. For instance, if you just went with a dual gang pot so that you get attenuation between stages 1&2 and 3&4 simultaneously, you might find that the way it cleans up as you turn it down "just works". If you have two separate knobs it is hard to simulate that dual pot action. Of course, I'm more or less assuming that your favorite distortion tones come with the gain pot fully dimed. In either case, you could probably make individual resistor adjustments at both stages to dial in your "sweet spot" at whatever rotation position you want. Get it set up so that at some knob position the circuit values end up exactly as they are now.
In my original response I was suggesting trying some slight local negative feedback around each stage to reduce the gain (10M-30M resistores, etc.). Might be just enough to scrub the edge off if you do indeed have "too much gain" as you mentioned in your original post. It's not something I have really tried, but I have taken note of it in a handful of amp schematics over the years.Not sure what u r asking RE: "the feedback"?Comment
-
Perhaps look at the splawn nitro design for inspiration. They're both 4 stage designs and everyone of them is bypassed! Key things to note here is that the first stage uses a large plate load but then has a divider right before the pot to get the signal down. The second stage divider dumps a lot of signal too.Comment
-
Because i never had it where there was a smallish load instead of a divider where the gain pot is and a gain pot where theres a load now. I just had both places with 1M gain pots. Anyways, tried it last nite with a 150k load where the gain pot was an used the gain pot at 2>3. Sounded like s*it. Now i know why all designs put the pot near the beginning.Originally posted by cbarrow7625 View Post
Ahh, yes, i tried that once. Had to change a bunch of 1 and 2.2m resistors to try it and i forget the cap value i used. But i think i copped it from a dumble. I seem to recall not noticing much of anything so i s*it-canned the idea.In my original response I was suggesting trying some slight local negative feedback around each stage to reduce the gain (10M-30M resistores, etc.). Might be just enough to scrub the edge off if you do indeed have "too much gain" as you mentioned in your original post. It's not something I have really tried, but I have taken note of it in a handful of amp schematics over the years.Comment
Comment