proper voltage at PI input ?

[Alan0354]

All that said is only hot air, I am not experience enough to say it is right or wrong to optimize the sound. Other more experienced people here can tell you whether it is correct to NOT clipping the power amp. Some people might like the sound of the power amp clipping. But you ask how to make sure you don't clip the power amp, so I show you a standard way we do all the time on a system ( not on a guitar amp!!!).

You can match the amp till the cows come home.......and then it may still sounds like crap!!!! These amp goes beyond theory and I am still treading on water!!! People here that have life long experience can give you more advice. Just the other day Chuck H and JMF wrote just two paragraphs commenting on my question, that was a light bulb moment.

[daz]

Where it become square? At the input of the PI, or at the output of the power amp?

speaker out

1) lower the gain of the power amp by changing the closed loop gain of the power amp so the power amp starts to clip at the input level where your master volume is at 10. OR

2) Use voltage divider before the master volume to reduce the output of the master volume so you barely drive the power amp to clipping when you turn the master to 10.

#1 i don't understand. #2 is what i already mentioned i've done. It helps, but i think i have even more level than i first thought. I may try even more attenuation before the master. Even with 70% signal reduction b4 the master the amp gets crazy loud at 10:00.

This is called matching the input and output dynamic range of the different stages. This is so simple and obvious that a lot of people over look it. This even remind me I have to do this myself!!!! I can talk a good game, I notice the sound of my amp change quite a bit when I crank it up. I need to do it myself!!!! Hope this help. Sure help me!!![/QUOTE]

[cyclone]

daz, I went through the exact same thing, If you don't put a voltage divider in front of the PI, a bag of ass is the result. I think I used 220k, 47k on my 3 stage, no cathode follower amp with a high gain PI. with a 1.2k and 47k like a Vox LTPI, the PI has a gain of 30! So, knocking the signal down before smoooothed things out. if it sounds good it is good Interstage attenuation is key. now that you have a scope, set the signals up so that the sine starts to flatten at the pi input at 6 or whatever and the pi starts to flatten out just after that)))))

[cyclone]

after I discovered that, I cannot understand why an amp without that attenuation sounds good, most don't or only sound good clean. well, being a caveman, my idea of clean is ac/dc rhythm guitar, which has some bite but isn't dirty, imho. 90% of amps out there weren't meant to be overdriven, sound like crap when pushed hard.Thats why pedals are popular, crappily designed, assbag bugzapper distortion stylings are the only high gain sounds available to the goateed triple nickleback rectified music store wankers.

[daz]

Well, i DO have a master before the PI but it's so touchy with all that level in front of it i get WW3 before the knob is even to 9 o'clock. I have a 2k cathode on the PI, but i tried a bigger tail and that kinda mushed out the tone. Plus i think it screwed up the NFB if i recall. One thing i think is making this hard to figure out is the room in which i mostly work on it....small, bright, and the speaker too....a 12L. So those naturally tend to make for a harsher tone when you turn it up, and in a small room it's hard to know how much of it is the room just being too small for that volume. Thats why i take it to a friend's house or to work when i wanna work on the loud stuff.

The preamp is great. I need to stop playing with that because it's as good as i think it's ever gonna get. I just feel somewhere in the PI on theres something thats gotta change but i haven't found it, just bandaids like this voltage divider in front of the master. But there is something happening in that 2nd half of the amp that isn't quite right i think. But then on the other hand, if i attenuate the signal b4 the master and scope the output, the wave stays the same all the way to max aside from just becoming more accentuated in it's shape. So i think it looks like thats good but still i feel the high volume tone isn't right. When i use NFB it goes away when i turn it up loud, so what good is it? I wonder if thats the issue, but then why does it do that.

[Alan0354]

speaker out



#1 i don't understand. #2 is what i already mentioned i've done. It helps, but i think i have even more level than i first thought. I may try even more attenuation before the master. Even with 70% signal reduction b4 the master the amp gets crazy loud at 10:00.

This is called matching the input and output dynamic range of the different stages. This is so simple and obvious that a lot of people over look it. This even remind me I have to do this myself!!!! I can talk a good game, I notice the sound of my amp change quite a bit when I crank it up. I need to do it myself!!!! Hope this help. Sure help me!!!

[/QUOTE]

#1 is just a different approach of #2. The gain of the power stage is governed by the NFB. If you ignore the open loop gain( forward gain) of the power amp, the gain is about (100K/4.7K)X2=40 minimum. So if you put 1V, you see 40V at the output of the amp( to the speaker). This is because you tap at 4ohm which has 1/2 the voltage of the speaker tap. So if you find you are clipping at 40V peak to peak at the output, you know the input dynamic range of the power amp is 1V peak to peak. You follow me?

So if you find out the output of the master volume at 10 is 4V peak to peak. Obviously you are going to over drive the power amp. So you lower the closed loop gain to 10 by changing the feedback resistor. You tap on the speaker tap instead of the 4ohm tap, you reduce the gain by half, you then reduce the feed back resistor by half, you get a closed loop gain of 10.

Now looking back, you know the clipping at the output is still 40V peak to peak. With the closed loop gain of 10 now, your input dynamic range is 4V peak to peak. Now you match the output dynamic range to the input dynamic range of the power amp.

BTW, this has nothing to do with 12AT7 or 12AX7. As long as the power amp is not clipping and able to feedback, the input swing is just COMMON MODE range. Only when the power amp is clipped, then the input range of the PI tube comes into play.

[Chuck H]

Daz,

You DO have a ton of signal hitting the PI, as Juan noted. The reason is that you have pumped up the preamp, as Juan noted. Right now you use a master volume after the tone stack to attenuate it to where it's all cooperating. And that's a reasonable way to do it. Asking how much signal should be at the PI WRT guitar amps is like asking how much BBQ sauce to put on ribs! Recipes vary. It depends on your goal. In your case you want the amp to produce full power and NOT create a ridiculous amount of power tube clipping. So... As mentioned, you can attenuate the circuit at the master volume to achieve this. My suggestion would be to add a series resistor before the master (not too high, that could add noise) and then use a smaller pot for the master volume. The series resistor will help keep the tone stack doing it's thing without interruption at different master volume settings. Do it like this:

Set the amp up for max clean signal from the preamp (use your newly acquired scope Put a 220k resistor on the wiper of the treble pot to the master volume. Jumper the two ungrounded legs of your master volume. Now turn up the master until your power tubes start clipping. You DO want to be able to clip the power tubes at least a little. Again, use your scope. Little flat spots on the wave is what you want. When you see that, stop. Measure the resistance on your master from the jumpered tabs to the grounded tab and replace your master volume pot with the nearest standard value to your reading. So, if the test results in a resistance of 220k, replace your master volume with a 250k pot (audio taper). If you get 90k or 110k use a 100k pot. That's all it should take and your amp will be tuned to the circuit you have built.

[cyclone]

a voltage divider is not a band aid, it is a gain matching device, there is a certain AC voltage that will drive your output tubes to max power. you can exceed it, but you should know what drive voltage you are looking for out of your PI for a given output tube type. otherwise, every other mod you make is a "band aid" because you haven 't designed your power amp correctly.

[daz]

The last two posts have my head spinning, but chuck, i get what you said basically and will try that. Just have to read it carefully as i try that. Might be a while b4 i can tho. I do feel like theres something else going on besides signal level, and I think thats what Alan is addressing. But it's way over my head so i'll have to re-read it slowly and try and comprehend it. The biggest obstacles in trying to understand much of what i read are words/phrases i don't understand like "open loop gain".
Thanks both of you....

[daz]

a voltage divider is not a band aid, it is a gain matching device

\Depends entirely how it's being used. If theres a problem that should be addressed a different way but VD relieves the issue partially, thats a bandaid. I feel it is only because even when i reduce the gain to a very reasonable level i feel the higher volume tone is still too harsh. But again, that may be my perception partially if not fully due to the room and speaker as i mentioned earlier.

[daz]

Chuck, it just occured to me that what you suggest is what i was doing today, IE: a voltage divider in front of the master. Same thing. Maybe you already realized that but i didn't. If i put a VD in front of the master, you have a series R from the treble to the master input lug, and also from there another resistor to ground. the net result is a series resistor, and the other one in effect lowers the pot value. So it's the same thing. I never tried such low values tho so i just did.I didn't do it with the scope as you suggested, just wanted to see how such low values would be different. So a 100k series and 100k across the pot to make it a 90k pot. I have no idea how this affects the high volume sound, as if i test that at the moment i'll be shot ! But it sure dropped the level a lot more than the values i used earlier, which should have dropped the signal a lot more than this. I guess it's due to the 100k vs 470k going to ground. Actually, with so many things going on theoretically there, my tiny brain can't follow. But the tone at low volume was even sweeter than usual it seems. Can't wait to try it turned up. (not holding out too much hope tho)

[JoeM]

daz, I'm admittedly confused by this. It seems some of what your saying is contradictory, but if I (at least partly) understand, your not happy with the 'range' of control at the master volume. If that's the case, why not do what was recommended here for the touchy volume on most Fenders, and what I did on my DRRI. Instead of the 1M pot, replace the 'top' of the pot with a 660k resistor and change the pot to 250k.

[daz]

daz, I'm admittedly confused by this. It seems some of what your saying is contradictory, but if I (at least partly) understand, your not happy with the 'range' of control at the master volume. If that's the case, why not do what was recommended here for the touchy volume on most Fenders, and what I did on my DRRI. Instead of the 1M pot, replace the 'top' of the pot with a 660k resistor and change the pot to 250k.

I've done that but thats only part of the issue and not the main issue. But changing the pot value alone doesn't help. Adding that series resistor too, PLUS using a 100k to change the pot value to a very low 90k seems great. Like i said, i can't turn up now to tell, but the tone is sweeter leading me to believe possibly something like oscillation or some unwanted distortion at the PI is relieved. The touchiness of the pot is far and away better too, but as i said thats not my main concern at all. Sorry if i confused you, but realize that with all that i have tried and all the replies i'm trying to answer only confuses things more. In short, i feel there is far too much gain hitting the PI and that the tone suffers even with the master low for whatever reason.

[Chuck H]

If i put a VD in front of the master, you have a series R from the treble to the master input lug, and also from there another resistor to ground. the net result is a series resistor, and the other one in effect lowers the pot value. So it's the same thing.

It's actually not the same thing. A VD in front of the master (That's "voltage divider". Not the "other" VD ) you add a circuit and do nothing for the interaction of the tone stack. Adding a resistance to the treble wiper and changing the master value is the simplest and most effective way requiring the lowest part count. I'm a stickler for eloquence. EXAMPLE: Most Mesa products could use a lot of "Muntz-ing" (Look it up. "muntz" and "TV". The guy was known for removing peripheral circuits that didn't make a difference in the macro picture. interesting and inspiring. But take it with a grain of salt). To be effective the attenuation needs to avoid affecting the TS and put you in the target zone at the same time with the lowest part count. The simpler the circuit the less noise and complication you add with peripheral circuit implementation.

[daz]

It's actually not the same thing. A VD in front of the master (That's "voltage divider". Not the "other" VD ) you add a circuit and do nothing for the interaction of the tone stack. Adding a resistance to the treble wiper and changing the master value is the simplest and most effective way requiring the lowest part count. I'm a stickler for eloquence. EXAMPLE: Most Mesa products could use a lot of "Muntz-ing" (Look it up. "muntz" and "TV". The guy was known for removing peripheral circuits that didn't make a difference in the macro picture. interesting and inspiring. But take it with a grain of salt). To be effective the attenuation needs to avoid affecting the TS and put you in the target zone at the same time with the lowest part count. The simpler the circuit the less noise and complication you add with peripheral circuit implementation.

Draw it on paper Chuck. Look....voltage divider before the master: a series R to the input lug, a load from input luig to ground. Thats a VD before the master, right? Now look at what it also does....the load, since it's ground to input lug is a R across the pot, thereby changing the pot value. You still have that series R there like you suggest. If that pot were a 100k instead of 1M and just had a 100k series in front of it, wouldn't the result be the same? Maybe i'm wrong somehow, but when i did what you say I looked at it and had exactly as i did before, a VD before the master. Only difference was the values i used. You mentioned a R from the in to out lugs, but only as a temporary R to determine the pot value i need for the level i want, right?