My version of tubenit's enhance cap

[12AX7]

I tried his idea but it didn't work for me. I also tried caps ranging from 100 to 500pf from plate to cathode on the gain stages, on the OTHER side of the PI, and across the PI plates where so many designs use a 39 to 100pf.  All this to try and smooth and round that nasty little spitting staticy high end at the very top of the frequency spectrum like tubenit was trying to fix and apparently did. But for my 2204 style home brew it didn't do anything good.

Then i tried something that made me think why didn't i ever try this before. Actually, i have, at least in a way. The common cut control used on voxes and now many other boutique amps. I didn't use that because i've done that before and like tubenit talked about, i always felt the blanket effect. But this time i thought instead of using a larger cap that will outright darken the heck out of the tone when the pot is all the way to that side, why not a small cap and no pot to do the same kind of trick they do on the PI plates. A cap small enough that it needs no pot because it only affects the very top like those other smoothing tricks. But hopefully w/o the blanket effect. So instead of a cut control a small cap in the same place, a 125 pf in my case. (2 250pf's in series)The only difference between this and the PI cap marshall and so many others use is that it is after the coupling caps from the PI plates instead of on the plates themselves. I guess it has something to do with AC vs DC, but whatever it is this did for me exactly what tubnit said his did for him. The top is now rounder and unoffensive. I still have all the top and jangle it had before, but the spiky staticy nastiness is gone and it's sweet. I only posted this because i don't think i have seen this before aside from the cut control which like i said was never a fix for me. the problem with things like that is that a guitar NEEDS a ton of high end to cut right in a mix, and when you try and get rid of that  staticy garbage you kill a little top end and the tone no longer cuts thru the mix and sounds lifeless because you've also removed a necassary component of the top. This allowed me to keep it all and just remove that garbage. I've looked for this for years and just happened upon this coincidentally soon after tubenit found his fix. Must be something in the air lately !  

Anyways, i could wake up tomorrow and find it sucks. But it sure seems like the real thing unlike anything i've tried yet. I'd like to see others try and and hear what they think. To me it seems too easy because after years of trying to fix this kind of issue i cannot understand why others like marshall and the tons of others that use plate caps instead haven't used this. Maybe when I plug in tomorrow i'll know why, but i sure hope not.

EDIT: make that 166PF. Just noticed one of the 2 caps i used in series is a 500PF. 500pf + 250pf in series=166 pf. Tried 250pf also and this worked well too cleaning it up even more, tho i'm not sure if it lost a tiny bit of necassary top. More evaluation needed on that one.

[Tone Junkie]

Could you draw a picture somehow Im not seeing where your talking about in my mind.
thanks Bill

[tubenit]

As I understand it, this is what he is saying works OK.

With respect, tubenit

[12AX7]

As I understand it, this is what he is saying works OK.

With respect, tubenit

Exactly. And i'm not sure yet about the value. But i think that 160 is probably just about right because while it could use a tad more smoothing, in a mix i think any more may not be best. I have yet to test it tho. I always check my work by cranking up some MP3's and playing along to see how well it fits in a mix and i'll be doing that this weekend. And i always find the more  high end, even that "hash", the better it cuts thru, even to the point where it has more than you may wanna hear playing solo. So for me the trick is to eliminate that hash as you call it to whatever extent doesn't compromise the sound in a mix and still cuts enough of it out. The 160pf eliminates it enough for me, but 250pf was better and i need to evaluate it in a mix to see if any of that needed top end was sacrificed. Like i said it doesn't kill or blanket the high end anything like snubbers and all those other typical applications. But it somehow seems to round off a bit of the top as the value gets up to 250pf or more, (best i can explain it) but it does not have that blanketing effect at all. In any case, whatever i use it has been a major find for me. maybe someone can explain why it works so much better on the DC side than the AC side?

[The_Gaz]

My understanding is that placing the cap before or after the coupling caps has the same effect...

[kagliostro]

For that I can see in that position, in some way there is a discharge to ground due to the 220K resistors connected in each side of the cap and to ground, the resistor's value isn't low, but sure they do something respect to the cap in that position

however that is the same of the Vox AC30 tonecut being the pot only active if turned to have resistance, the same of this cap connection if turned all resistance off, the pot gives only a way to vary the efectiveness of the cap, something like if you have more than one cap with different values, so we can define this like a fixed tonecut

Kagliostro

[Geezer]

The AC30 has a .0047 cap (w/ the pot), which is a whole different freq range being affected.

The small >200pf caps are only trimming the very highest frequencies, the "undesirables" I suppose.

Another tool in the arsenal of tone adjustments....

G

[jojokeo]

The small >200pf caps are only trimming the very highest frequencies, the "undesirables" I suppose.

Another tool in the arsenal of tone adjustments....

Always nice to pop in and find another golden nugget and it's the subtle or little things (after all else) which can make all the difference in the world. Just ask your wife or signifigant other about this fact. When this happens I feel like Corneilius the prospector throwing my axe up, licking, and instead of "nothin' it's somethin' " afterall. (Rudolph TRNRD referenece - 'tis the season) Good job and thanks for sharing even if it's a modified "old technique". You got the nasty teeth out of that bumble-ing high end, it's a humble bumble now.

Happy Holidays all- Keo

[12AX7]

The AC30 has a .0047 cap (w/ the pot), which is a whole different freq range being affected.

The small >200pf caps are only trimming the very highest frequencies, the "undesirables" I suppose.

Another tool in the arsenal of tone adjustments....

G

Right, and thats what i was saying. (probably not as well tho) I hope someone tries it because it's always nice to know it's not my ears playing tricks. this tweaking business CAN and often is very tricky. But i must say this is the first time i ever found a high attenuation cap tweak that i didn't begin to hear it's weakness after a few minutes of playing. It's one of those thing i usually evaluate correctly within a very short time. This one has held it's ground so far against my extremely nit picky ears.

My understanding is that placing the cap before or after the coupling caps has the same effect

Not in this instance, not for me. I have tried putting them from plate to plate god knows how many times, each time out of frustration hoping to hear something a little better each time i've tried and it never has been anything but that blanket on the high end in a big way. Not even a maybe. This one is a "whoa" moment. It's nite and day, so i'm not sure where you heard that but give it a try and see if you get the same result. And again, i have no idea what the theory behind it could be, but it IS AC vs DC. So somehow therein lies the answer i'm sure.

[kagliostro]

My reference to the AC30 Topcut  was only meaning that this new approach is a refinement in the use of an old technique

I'm still thinking why there is difference of result due to the position of the cap before or after the other two standard caps, but I'm not smart enough to find a reason

surely those who have discovered this thing is to be thanked for sharing it

Kagliostro

[Tone Junkie]

Thanks Tubnit for putting it out there in a drawing its usually late when Im looking at things. Brain is not working well buy then.
 Got a swing shift job at boeing kind of cut into my amp building time, odd hours i look forward to trying these out just about have a trainwreck done.
12ax7 thanks for sharing with us.
 Bill

[12AX7]

No thanks needed, thats what forums are all about. But please share your thoughts on it if you try it.

EDIT: for the record i have settled on 100pf. Seems to be all i need to remove the nastiness and leave the top end intact.

[jojokeo]

The way I see it or analyze it is the only way to truely filter high end of the "undesireables or hash" as it is described while not simultaneously affecting the high end frequency response of the signal is through specific filtering similar to what a Noise Filter does in a high gain signal chain. This would mean sampling, then filtering, then actively replacing the high end response back to the original signal. Any filtering w/ a capacitor alone is passive filtering and it cannot be selective in what it removes and what it doesn't. It will remove to ground everything above it's cut-off point and corresponding value.

Every point in the circuit that we include "smoothing" caps we are dumping high end. Each place in the circuit creates it's own influences on the product or signal "soup" as a whole. Each gain stage can be adjusted and/or modified to decrease or increase frequency range. In this case it is in the very last final part of the preamp or beginning of the power section immediately before the power tubes - whichever way you choose to think of it as. It is the final point of the circuit to tweak anything before the power tubes do their thing and amplify the heck out of what we have created with all of the good, bad, and complexities in between.

I will say that kagliostro's comment earlier regarding the cap AND the resistance combination does indeed have merit as to how this makes 12ax7's opinion and "discovery" to an old circuit technique valid. Where it is placed makes this a truely passive, analog, first-order high-pass filter. It is defined as the product of the resistance and capacitance is the time constant which is inversely proportional to the cut-off frequency at which the output power is half the input power. Fc = 1/2╥τ which also is Fc = 1/2╥RC. Therefore it will not remove to ground everything above it's cut-off point and corresponding value. This is why not all of the high end is affected and "tone blanket" is not smothering all of the high end.

A gain or volume pot's treble bleed is a perfect example and application of where I like and prefer to use the cap & resistor combination over a single cap alone. It does make a noticeable and positive difference to me where a cap alone causes all of the frequency above this value to pass making the control too bright in lower settings. Not having anything there and the tone becomes too dull in the lower settings. Finding the "right" combination makes it perfect.

[stevehoover]

Just guessing.... (I know enough to be dangerous) isn't this about the same spot in the circuit that fender used a 100- 150PFpf cap to dampen Hi-frequency oscillation
in the 5A3 - 5B3?

Steve

HAPPY HOLIDAYS EVERYBODY!

[12AX7]

I just looked at both those but only one had it, and it was on the plates not after the coupling caps.

[12AX7]

Man, I must be spacing because i just noticed a few marshalls use it like this, Including the 2204 which is what mine resembles closest ! I looked at a few others and sure enough most seem to have it on the plates. I suppose i never noticed the 2204 has it because i have always ignored those caps due to the fact that they always dulled the top end to me. I guess when i looked at the 2204 schematic in the past i just didn't pay attention so never noticed the slightly different placement which makes all the difference in the world. They use a 47pf tho which i found too small to really make enough of a difference.