Smoothing "Enhance" cap

[tubenit]

I was continuing to experiment with my SoLow Watt amp to see if I could make it even smoother when I discovered a mod that has really worked out well for me.

So I have found a mod that I have never seen on any schematic before.

I was thinking about "is there a way to add another smoothing cap" without muting the tone?  My thinking was the triode of the ECL84 was perhaps chimey-er than what I like.  So I decided to add a smoothing cap across the plate resistor into the LTPI.

It works great!  In fact, it worked so good, I tried it on the Tweed Overdrive Special also.

On the TOS, there was NO loss of the great tone features of harmonics, sustain, touch sensitivity, blooming, etc.......   Similar on the SoLow Watt (except the SoLow Watt doesn't bloom).

In contrast to this, I've tried other smoothing cap techniques that could just kill the tone and give it a muted "blanket over the speaker" tone.  This does NOT do that. An example of the tone muting thing (IMO), is the 10p cap across The LTPI plates.

However, it removed 90% of the remaining high end "hash" that was there. (I had previously concluded that wasn't possible with high gain amps). There wasn't alot of high end hash with the TOS but there was some noticeable on the SoLow Watt.

The high frequency notes are clearer sounding with no loss of sustain, harmonics, etc......   So now both chording and lead riffs sound even more articulate, IMO.

The most interesting feature is that the "enhance" cap seems to increase the attack on the note when playing fast riffs sort of like a sustain/compressor pedal does.  This was most noticeable on the SoLow Watt but it did the same thing on the TOS. So playing a fast riff, each note sounds out more pronounced and clear. I am wondering if the reason it does that is that removing the high end hash allows the notes to sound more clear and articulate?

I tried caps from a range of 120p to 390p on both amps. I tried Silver Mica and two different brands of ceramic caps.

On the SoLow Watt:  I found the 150p and the 220p Silver Mica caps to sound best. Smaller was too thin. Larger did not sound as "musical" to me.  I settled on 220p Silver Mica. Ceramic caps did not sound good on this amp.

On the Tweed Overdrive Special:  I found a 220p Silver Mica and a 251p Ceramic to sound best. The 390p Silver Mica sounded pretty good also.  However, the Ceramic 251p that Hoffman sells sounded the very best to my ears. It was smoother sounding and had more of a vowel tone. The 220p Silver Mica had more chime to it.

Summarizing what this mod did:

1)  removed remaining high end hash with NO loss of harmonics, touch sensitivity, sustain, blooming or other nice
     tone features

2)  it increased the clarity of the notes while adding smoothness to them. The notes have a more musical "sweetness".

3)  it increased the attack of the notes on fast riffs sort of like a sustain/compressor pedal does


IF anyone else tries this ......... PLEASE post the results even if they are different (positive or negative). I'd like to know if this works on most LTPI amps.  I used the same guitar and speaker on both the TOS and SoLow Watt, so I don't know if that factors in.  I do plan to try it on the Tweed BluezMeister and will let you guys know.

With respect, Tubenit

[The_Gaz]

Hi Tubenit,

Why on only one plate resistor? This seems like it would cause some weird phase issues in the upper frequencies you're cancelling out on only the input side of the PI. You realize that putting a cap across the plate resistor is exactly like putting the cap to ground, right? I'm also surprised you say a 10pf resistor across the plates sounded like a blanket. To be honest, I think that's an exaggeration, and I don't even think most people (well, anyone actually) can tell the difference between 10pf and no cap if there were no supersonic oscillations playing havoc. Either way, I'm gonna try your mod later today and report back.

[tubenit]

Why on only one plate resistor?

The truth was mostly just an arbitrary decision without any real foundation or understanding behind it. The (completely unfounded) reasoning is the signal path goes right into "that" triode of the LTPI. So I tried "that" one.
 

I have tried the 10p resistor on maybe 4 or 5 amps and removed it every single time. I can hear a difference. It kills the high frequencies, IMO. I don't hear that muted sound with other smoothing cap methods.  Hoffman refers to this same issue in his Library of Information with AB763 Fender amps.

This seems like it would cause some weird phase issues in the upper frequencies you're cancelling out on only the input side of the PI.

I don't care if it's causing weird phase issues or not. It sounds significantly better on all 3 of my amps. Again, it removes the high end hash/fizz without removing the high end frequencies to my hearing.

Try it. Honestly, if it doesn't work or doesn't sound as good ........ please report that. I would like to know if this is applicable for most higher gain amps or not?  I am happy with it & plan to leave this "enhancing" cap in.

With respect, Tubenit

[tubenit]

OK, I tried this on the Tweed BluezMeister.  Same results. I was somewhat skeptical that it would make much of a difference but it did. I thought the clean on the TBM was as good as I'd ever get on one of my builds. 

However, the enhance cap made the high frequency notes sound clearer. I would have stated the clean on the TBM did not really have any high end hash until I tried this cap.

Same results:  1)  clearer tone with NO loss of harmonics, blooming, sustain, touch sensitivity, etc.......   2) somewhat quicker attack on the notes when playing faster riffs  & 3) it added a sweetness and more musical higher frequency.

I hope anyone who tries this will post their results.

On this amp I tried 120p, 150p, 220p, & 390p Silver Mica.  I also tried the 251p ceramic.  And a thick blue enamel 250p ceramic of some type that I am not familiar with.

The amp sounded better with 120p, 150p and 220p Silver Mica.  The 251p ceramic didn't sound that good. The thick blue enamel ceramic 250p cap had ALOT of chimey tone to it and really sweet harmonics but wasn't as smooth.

I settled on the 150p Silver Mica.

With respect, Tubenit

[jeff]

 "Why on only one plate resistor? This seems like it would cause some weird phase issues in the upper frequencies you're cancelling out on only the input side of the PI. "


Just a thought but maybe one is all you need. If the one cap is shunting out the fizz in the first section of the PI and the second section is being fed by the shared cathode, maybe it's signal is already modified by the cap across the first section and there's no need for one across the second section's plate resistor??????

Make any sense? I don't know if that's right but that's my guess???
What do you think?

[JBP]

Got my SoLow Watt up and running last night . I check in every day to see what T has come up with so I was pleased to see somthing to tryout. I'm a semi-nubie so I don't understand how all this stuff interacts. I can confirm that this mod changes the attack and tone seems to be clearer. Might be nice to make it switchable. Really liking this amp. Thanks for the ride T

[firemedic]

I wonder if it would work on a cathodyne.....

[The_Gaz]

Firstly, I could only experiment with this mod today with the amp's PPIMV turned low (noise restrictions). I thought this would be a pretty good test, however, because the amp I tried it on can get very buzzy due to the PI.

The amp I tried it on was an AC30 with Hot switch, and I thought the effect was very subtle with 220p. I was really listening for the smoothing quality mainly, and heard a slight reduction in buzz, but nothing very different compared to other high frequency shunting methods on the PI. Trust me I wanted to hear it, especially because this amp is still quite buzzy with the Hot Switch, as you can imagine.  I heard nothing phasey going on, which was good, but couldn't hear anything significant until I went up to 470p.

I don't know if I'll use it, but I'm following this thread with great interest because I really wanna know what the caps doing electrically since it only on the input side of the PI, and if there's a good explanation for what you're hearing. I've been in amp building message board game long enough to know that one should never to tell a guy what he hears or doesn't hear. On that same token, I know that there's no magic in electronics :)

[The_Gaz]

I wanna clarify one thing with you Tubenit. You say that the mod removes high end "hash", but does not remove harmonics. What is hash exactly then? I'm not splitting hairs here, but the mod does remove harmonics, because besides noise, there's nothing else to remove! If you are talking about noise that's one thing, but I don't think you are. The reality is that it's removing harmonics, but I think they are just harmonics that are unusable and unmusical to guitar. Do you really mean musically relevant harmonics?

Although your descriptions of your amps and mods are usually adjective-laden, I've always appreciated your consistency, and especially how you tried to quantify your experience with your table (don't have a link to that thread). But I think there's a confusion on my end between what you mean by harmonics.

[tubenit]

Gaz, Thanks for posting the results of your experiment!

What I mean by high end hash is the high end sounds fizzy or distorted rather than clear.

To give a visual description ............  it would be like me trying to read something without my glasses (blurry to look at) & then putting my glasses on. Doesn't remove the print. It just puts it in focus.  

To my ears, it's not removing high end frequencies but making those frequencies clearer & less fizzy in tone.

Negative feedback can sometimes clear up a tone. Maybe this is somehow functioning in a similar audible way?

I A/B'd the amp with and without the enhance cap on each amp. I could never hear a loss of harmonics. Instead it seemed to sweeten the tone with the hash removed.

With respect, Tubenit

[topbrent]

I posted this schematic over at AG, but thought it would be helpful here also.

Mesa uses a similar arrangement on the Lonestar amps.  Mesa's implementation differs in which "side" of the PI plates has the cap tacked on.  Theirs is opposite of yours.

*120pf across the 100k PI plate.

[tubenit]

WHAT !!!!!??????   Are you telling me Mesa Boogie already stole my idea?
 

Seriously, they put it on the "wrong" side. The way they did it will totally throw off the parabolic flux of the emminence harmonic convergence. No way that will sound good.

Besides, I think (?) their cap is called a MBCC (Mesa Boogie "corrective" cap) which sounds so rather 60-ish and passe' and dysfunctional.

Mine is called a TEC cap.  Therapeutic enhancement cap that sounds much more positive and affirming (ala Oprah and Dr.Phil).  Plus just the acronym of TEC sounds robust and more modern .......... don't ya think?

 

As I have said before ............ anytime I think I've actually come up with a new idea ........... I find someone has already done it.  Oh well!

Hopefully, at some point, Mesa Boogie will realize their mistake and switch the cap to the other LTPI plate where it should be.

 

With respect, Tubenit

[John]

I was tinkering on mine yesterday, wish I had seen this thread first. When I try it out, I'll be sure to use the correct way... I'll 'nit it!!
Thanks again for sharing all your ideas.

[phsyconoodler]

What about Dumble's cap from plate to cathode?They use from 270pf to 500pf.I tried my amps without them and the high-end hash is there.with them it's quite a bit less.Same effect?

[firemedic]

It seems like tubenit's version bleeds treble from the "push" side only, leaving the "pull" side unaffected, as the cathode signal should not (?) be affected by a bleed cap across the plate resistor. Mesa does the reverse. A plate-to-cathode cap would be local NFB. And a cap across the PI plates would be a different NFB. 

Unless I am mistaken?

[Geezer]

What about Dumble's cap from plate to cathode?They use from 270pf to 500pf.I tried my amps without them and the high-end hash is there.with them it's quite a bit less.Same effect?

The Dumble caps are on the preamp tube (or overdrive section) plate resistors.

Tubenit's cap is placed @ a different location in the circuit (across the plate resistor of the "input" side of the LT phase inverter). I don't think Ive ever seen a Dumble (or any other amp, for that matter) place a cap in that location(?)

I'll be experimenting with this within the next few days & will post results.

Geezer

[tubeswell]

It seems like tubenit's version bleeds treble from the "push" side only, leaving the "pull" side unaffected, as the cathode signal should not (?) be affected by a bleed cap across the plate resistor. Mesa does the reverse. A plate-to-cathode cap would be local NFB. And a cap across the PI plates would be a different NFB. 

Unless I am mistaken?

The plate bypass cap 'shunts' some 'highs' back into the B+ rail. These highs will no longer appear at the plate coupling cap. Since the plate swing on the inverting side determines the signal swing at the cathode in an LTP, then the signal at the cathode would also have less highs in it - and this feeds into the non-inverting side.

[PRR]

> A plate-to-cathode cap would be local NFB.

Not really. There's voltage gain but no current gain. Unless the NFB part has current gain (a step-down transformer) there won't be any real NFB.

It's just loading the plate (the cathode is, relative to plate, "ground", near-enough.)

> And a cap across the PI plates would be a different NFB.

This again is just loading.

> Since the plate swing on the inverting side determines the signal swing at the cathode in an LTP, then the signal at the cathode would also have less highs in it - and this feeds into the non-inverting side.

Your analysis sounds good, but....

The cathode impedance of a triode includes plate load divided by Mu. The cap shunting the plate load DEcreases front-side cathode impedance, which forces more highs to the back-side cathode.

I was not sure of my analysis so I fed it to the Dummy. SPICE can give very wrong answers, but if the question is framed correctly then it often gives a right answer with 8-place resolution and pretty pictures.

The front side declines as expected.

The back-side _rises_ in the highs.

In push-pull this would change the overload harmonic structure from odd to even. (The Presence network also does this, which is why I omitted it.)

With the suggested ~250pFd cap, the difference is very small over the entire e-guitar range. I don't "see" a difference which would make me mod all my amps. I accept tubenit's ear-observation; just can't understand what is going on.

If it were just one amp, hypersonic instability could be a reason for large change of strained tone with small change of treble response/phase. However he's found this to be good on several amps, which can't all have borderline hypersonic instability.

[12AX7]

I tried this with a 2204 style amp that has a preamp thats a little different but the power section is about the same. Tried it on both the 100k and 82k and used a 250pf. To me it just sounded like the typical cap across a plate resistor. The highs were slightly reduced just as that would do in a preamp stage. I prefer the amp without it. I thin if i read PRR's post correctly he's suggesting it may be fixing a issue rather than a tone tweak. Ultrasonic oscillation or  whatever. To me it was something some people may like even with what it did to mine. But i am accurately aware of what certain tonal aspects do to my tone in a mix and i know the bit of top it removed would seriously hurt the amp's ability to allow the tone to cut thru. I have found i need that spikey top that may not be great solo, but in a mix it helps keeps the tone from getting lost.

I'd like too share something i have found lately too. I always used 1 or 2 RK networks (which i think they're called) like in a JCM. The 470pf and 470k in parallel to allow for a bit more of the top end to pass in relationship to the rest of the range. Well, i started experimenting and found that if i added a few of them it was a case of the more the merrier. In the gain stages, and especially near the beginning it really did some serious magic in giving the tone considerably better articulation and harmonic complexity.  Plus i used a 250pf on 2 of them and a 470pf on the other. So the values can make a considerable difference too. The 270's added sweetness without adding too much high end. The amp improved to my ear dramatically. I always used these but never more than 1 or 2 and apparently never in quite the right places. I have 2 now plus the one that is on the gain pot which is always down around 12:00 to 2:00. So theres some of that always going on there also along with the other 2. All are within the first to second stage. Probably one of the best tweaks i've found, and amazing how adding just one more made such a big difference. But the key is finding where to put them because it did not give me such great results later on in the gain stages. Anyways, i think in amps with cascaded preamps the use of several of these when in the right place and the right values can work real magic. It was one of those tweaks where you just know you hit on something big and something very very right happens to the tone that you have long been trying to accomplish.So simple, bit thats the fun of building and tweaking. Like panning for gold and finding that one huge nugget.

[Geezer]

Had some time to try this & here are results, observations & general thoughts.

The amp is my Maxim Triple Drive (basically a HoSo56 for the "clean" channel, cascaded into a triode gain/CF (Dumble "ripper"-type circuit) for OD.

I tried a wide selection of different types of caps (ceramic paper, ceramic plastic dipped & silver mica dipped) and values (68p, 100p, 120p, 180p, 220p, 250p, 330p, 390p, 470p, 590p).

The effect of the cap is subtle, but significant to my ears, & "fixes" a problem that can be difficult narrow down in these high(er) gain circuits that generate most of their distortion in the preamp (vs power amp distortion). They can tend to have a certain "buzziness", and many solutions have been tried, some working very well (smoothing caps across preamp plate resistors, high capacitance sheilded cable, high value grid stoppers, etc) & others, not so well.
In my amp, I had fine tuned the circuit to as close to "perfection" as I could, but there was a certain "something" that was still not quite pleasing to my ears on certain notes, especially on higher gain solo notes in the upper register. It was almost akin to & sounded similar to what is commonly called "cone cry" in some speakers, but not quite the same, & was present equally with different speakers, both new & old & different types & manufacturers. I have heard some builders refer to it a an "artifact" riding on the note.....I refer to it as "hash".

These caps ALL had an effect on removing that bit of "hash" & giving the finishing touch to an already "perfect" amp, but some values had a more pleasing effect than others.

One interesting thing is that the value of the cap has a slightly different effect on frequency response when used in this location (PI input side plate resistor) vs the normal use across the preamp Rp.
In the preamp, the cap simply removes more high frequencies as the value of the cap is increased (blanket on the speaker effect).
In this application, the cap value seems to actually shift the frequency response (not simply add or subtract higher frequencies). The effect seemed to be on the upper mids, and affected the "vocal" qualities of the amps "voice" (if that makes any sense). The high frequencies were still there when the cap was added, but they were "smoother" & "clearer" (less unwanted artifacts/noise on top of the notes), and increasing or decreasing the cap value (within reason) didn't have as much effect on this desired result (smoother high frequencies) as it did in changing the upper-mid "voice" of the amp.

I tried all the above mentioned values & types, settling on 180p as the best value (to my ears & for my amp). None of the values were "bad", just "different" & not what I was wanting to hear from this particular amp.
I had both dipped & non-dipped (paper covered) ceramics in the 180p, & finally settled on the plastic dipped unit as the "best".

Again, this is a subtle effect (to my ears), but is quite useful in putting the "icing on the cake" of an already good sounding amp.

I like it & the 180p is staying in the amp......Thanks Tubenit, for the new tweak!  

$0.02

Geezer

[LooseChange]

Thanks for all the info...

Question: How does this effect the clean mode of the amp?

[tubenit]

How does this effect the clean mode of the amp?

Exactly the same. When I tried all the different caps to select one, I did the experimenting on the clean channel and then tried it on the OD channel to see if it sounded good on OD also.

That was with the TBM and TOS amp. SoLow Watt does not have a clean channel.

with respect, Tubenit

[Geezer]

How does this effect the clean mode of the amp?

Exactly the same.

Dittos.

[The_Gaz]

Ok, I was able to do some tests at full volume today. I was able to draw the most striking conclusions while testing with clean and semi distorted tones. The amp I tested with is almost identical to an AC30 Top Boost.

I agree with Tubenit that the cap does not give the 'blanket over the speaker effect', and I think this is because it does not actually alter the very high frequencies (at least to the degree that a cap across the PI plates does), but rather the very high-mid frequencies.  This seems in-line with what Geezer heard as well.

In an AC30 TB, whose tonestack is already very scooped, the 150p cap made the treble range sound thinner, and switching to a different value seemed to shift scoop down in frequency. I think the 'shift' I heard when adding the cap and switching different values is also in-line with Geezer's observations, however I do definitely hear a reduction. (I'd actually argue Geezer does too - How can you hear a 'shift' in frequency if there's no addition or subtraction?)

The particular amp I used has a typical tone stack lift switch as well, which gives a boost in upper-upper frequencies. In this mode, it was even easier to hear the effect of the cap. The cap made the amp sound subjectively better because those frequencies were reduced slightly with the "enhance" cap in place, making the tone clearer, and less spikey in those frequencies. I  really can't say how the effect would sound in a band context, but in isolation sounded more pleasing.

I did a lot more tests with different settings that I won't go into, but sometimes the cap sounded better and sometimes worse. I don't think it's as simple as a better or worse mod, as with most mods. The effect is definitely unique, and I'll add it to my toolbox, especially because that upper-mid harshness is tricky to get rid of outside the tonestack modifications.

What's most intriguing is that, to me, the mod seems to be neither a low-pass or shelving filter, but kind of a notch filter. I'm really curious to know why it has this effect, and I'm guessing it has to do with phase cancellation, since the cap is just one one said, but of course, that's just a presumption.

[Tone Junkie]

Cool I cant wait to try it in my amp.
Thanks Bill
PS first I will build another cap switching box with the right values, any excuse to build another tool.   

[frank57]

Is this a similar  idea to what Marshall and Soldano
do in their amps with a cap across both plates on the phase splitter
after the resistors?

[Geezer]

No, that's a "snubber", also found on tweed Fenders (where Marshall & Soldano got the idea). It's for curing possible oscillations @ the PI.

[tubenit]

I think I will officially call this "enhance cap" the TEC cap.  

DaGeezer (great guy that he is)    put me on to trying lowering other smoothing cap values AFTER installing the TEC cap into my TOS.

Experimenting, ......... I discovered that smoothing caps that are 500p or larger across the plate resistors add some slight middle tone grit/grind that I don't like. They do effectively remove the high end hash but they add the slight grit in the mids unfortunately.

Well since the TEC cap effectively removes the high end hash, I am able now to lower the smoothing caps on the 12A_7 tubes and the 5879.  In fact, I removed the plate resistor caps (.001)  altogether and instead add the more typical plate to cathode smoothing caps using a 390p instead. It also allowed me to up the cathode cap from 2.2uf to 4.4uf on the V2-3. That gave the amp a nice mid boost without grit.

Overall the amp sounds clearer, somewhat brighter & more expressive. Some improvement in harmonics. Slight loss in the vowel/wah tone.  Overdrive and sustain is the same.

I did try a 2nd TEC cap on the other LTPI plate resistor. Used a similar value and it did NOT work at all. Unlike the TEC cap on the input plate resistor that enhanced the tone .............., the 2nd cap took away from the tone and took some of the expression out.

I tried about 5 or 6 other tweaks that were useless or lessened the good tone and immediately reversed them back.

This is the most tweaking I've ever done on an amp & I feel like it's been quite worth it.

Just thought I'd share the results.  Here is what I have done with the TOS to get the best amp tone I've ever gotten.

With respect, Tubenit

[John]

Thanks for sharing and documenting your work, as always!

[tubenit]

Just an update with some additional thoughts.  

My Tweed BluezMeister (when using a delay pedal) had some upper high frequency hash when the amp was somewhat cranked.

I added the "enhance cap"  82p after the LTPI coupling caps. It both eliminated the high end has and it added what I believe is a significant amount of "blooming" effect to take place.

This TBM has an EH  12AY7 in both V1 & V2 & an 12AX7 LTPI going into EH 6L6's. So, I am impressed that I can get this sweet blooming tone going on using 12AY7 tubes in those positions.

My theory is that an enhance or smoothing cap may (at times) be eliminated some mild blocking distortion/oscillation and then allowing a purer tone to come thru.  

Sort of like throwing a spinning top.  Prior to the enchance cap ........... the top wobbles & doesn't sustain very long.  After the enhance cap  (& with oscillation eliminated),  the top spins smoothly and sustains that for a longer period of time.   This analogy may not be accurate from an electronics standpoint, but it is a reasonable description from an auditory viewpoint, IMO.  

Good news, is that I am finding by trying out numerous different sized caps that I can usually smooth the tone out withOUT any loss of high end frequencies that I can tell.  IF you use too large of a cap, it will kill the tone and high end too much, IMO.
However, I am finding by trying out different ones, that I can typically eliminate high end hash without eliminating the tone I want.

I also have found that it is important to have a variety of both ceramic and silver mica caps to experiment with. Sometimes a ceramic sounds better ............ sometimes a silver mica does.  Not sure why that is the case?   ( I try to keep on hand:  47p, 68p, 82p, 100p, 120p, 150p, 180p, 220p, 250p, 390p & 500p).

The caps to cathode & the LTPI "enhance cap or TEC cap" is what I apply first.  Then I look to the other caps  (noted in green) if the smoothing is still needed.

With respect, Tubenit

[Geezer]

Sort of like throwing a spinning top.  Prior to the enchance cap ........... the top wobbles & doesn't sustain very long.  After the enhance cap  (& with oscillation eliminated),  the top spins smoothing and sustains that for a longer period of time.   This analogy may not be accurate from an electronics standpoint, but it is a reasonable description from an auditory viewpoint, IMO.

That's a great "picture" to help visualize the effect!

Sort of the same as what the Dumble guys do (using a totally different method) by balancing the PI plate voltages with a trimmer.

Thanks for the updates, T!

G

[punkykatt]

I wonder if it would work on a cathodyne.....

GOOD QUESTION:  anyone try it yet?

[nandrewjackson]

I'm gonna try this on my JetCity 2112RC (Astroverb).

I've tried the preamp caps across Plate R, and it just takes the high end character and flushes it down the drain.
I'll try the PI plate(s) with cap(s) and post here with results.


I really don't like the 475K parallel 475pF in the interstages , (marshall) with high gain.
They sound good with a 1960 4-12 cab or something similar, but try a open back 2-12 and you'll find the low end is absolutely hideous, (my opinion). 

My JetCity is half open back 1-12, and it comes stock with those 475K/475pF interstages, and some preamp plate Caps, I had to cut a lot of that out, again, personal opinion, but without a fully closed 4-12, it just isn't right.

[nandrewjackson]

+1 for the TEC therapeutic enhancement cap.

installed on "left" side of PI in JetCity combo.  150pF parallel with the 82K plate R.

Much better result than Picos in parallel with preamp plate R's.

It's not so much cutting the highs as smoothing them out. (as per my ear to typing fingers meter)

With the PI being part of the NFB from output transformer (presence control on JetCity) and only having the cap on 1/2 of the PI, I'm sure the complexity of the theory of what's going on is very high.  But, it sounds good, doesn't compromise stability or performance, and surely doesn't drain the high end like putting the cap on a preamp tube plate R. 

It's staying.

[jjasilli]

I've seen this done on preamp stages, per a prior post, but not on the PI.  Usually it's done to tame squeal.  The PI is an interesting location.  I think this:  every time signal goes through a preamp stage, a new set of harmonics is self-generated by the amplification process.   The new harmonics in one stage are seen as fundamental notes in the next stage, so the harmonics generate more harmonics, and so on.  By the time signal gets to the PI, it may be laden with lots of harmonics not present in the original guitar signal.  This treble bleed cap attenuates some, but not all, of these harmonic layers which built-up in the prior preamp stages.  Hence the "smoothing" effect, while treble is preserved.

I suspect the same result could be achieved in other ways with cleverly fine-tuned local or global NFB.  But this approach is simple & elegant.

[12AX7]

I guess i'm the odd man out here, but i have mine on a switch and after going back and fourth for several days i'm not hearing what other are, that is to say a difference between this and the typical cap across the plates. I DO hear the difference it makes, but to my ear both plate to plate and enhance scenarios simply cut highs which sounds like they're smoothing the highs because they're removing what you may perceive as harshness which is to my ear just a more extended range at the top of the treble range. I still can't seem to say for sure whether it's better with or w/o. I find that with the cap i can add highs via knob twiddling to make up for the slight loss, or w/o the cap i can cut them. But so far i can't determine that one is better than the other. But i definitely don't hear a difference i could describe as smoothing any more than if i cut the amount of presence or treble a tad. This to my ear at least would seem like extreme cork sniffing, and i'm always accused of hearing things that would be stupid to even consider like the nut's effect on fretted strings.

My only idea from why i'm not hearing the same thing as others is maybe my amp simply doesn't need it. It may be amp dependent for any number of reasons.

[tubenit]

Maybe it does depend on the specific amp?  

The typical Fender set up with snubbing cap just sounds like I put a thin blanket over my speaker and kills the highs.  The enhance cap smooths the highs removing the high end "hash" without muting them and allows the notes to bloom better on my amps.

Don't know what else to say?   :

With respect, Tubenit

[plexi50]

Pretty cool fine tuning adventure tubenit. I'll have to try this. I know what you mean by high gain hash in some of the amps i have had. Being on the razors edge of shrill

[12AX7]

What i found last nite playing the amp in a mix and switching back and fourth from cap to no cap is the same thing i typically find if i cut highs to any noticeable degree when playing in a mix. It cuts better and the details and harmonic complexity comes thru better. BUT, playing it by itself no music to have to cut thru, the cap sounds better. But it IS  a subtle thing and i think that i can pretty much approximate the cap by cutting highs a bit via the presence knob. So i'll leave it on the switch for a time. But chances are i'll end up removing it. It probably does depend a lot on the amp, and if as you seem to elude to, you are using this on fenders, i can see how that may be more needed than on a MV marshall type amp. Fenders certainly can be painfully bright.

[strictlyamateur]

This is fascinating stuff, and very important. I'd like to try this on an audio amp - the Jolida 102B - would I do that by putting a 220p silver mica in parallel with R7 and R8 in this schematic? Thanks!

https://el34world.com/charts/Schematics/files/Audio_pa/JD102B_2007.pdf

[tubenit]

I would try just R7.  I usually end up with around 220p in this position but you can try anything from maybe 120p to 390p.

You can also look at how Dumble style amps use "smoothing caps" from like pin 1 to pin 3 ………… or pin 6 to pin 8 on 12AX7 tubes.

With respect, Tubenit

[sluckey]

This is fascinating stuff, and very important. I'd like to try this on an audio amp - the Jolida 102B - would I do that by putting a 220p silver mica in parallel with R7 and R8 in this schematic?

If this amp is being used as intended, ie, to reproduce hi-fi music, then don't mess with it. If you are using it as a guitar amp, give it a try.

[strictlyamateur]

Thanks, Tubenit and Sluckey. Yes, I use it for audio. This thing has some harshness in the highs that makes it worth trying; I've fiddled with enough parts and components to think the problem is in the amp. I found this page in a search for a fix for EL84 amp harshness - Tubenit certainly has found a unique answer. I ordered some silver micas from Parts Connexion and will let you know (and cross my fingers).

[shooter]

I've fiddled with enough parts and components

I'm a neophyte - audiophyte, I've found source audio can be pretty harsh at the highs, even flabby at the bass.  I like Frank Zappa doing some Bach for my "test pattern"   

[strictlyamateur]

True, Shooter. My next move is a Schiit Loki equalizer if this doesn't work (or a solid state amp like Odyssey or Burson).

[strictlyamateur]

I've tried to do this a few times but there's not enough space in my amp to try it, a big cap and an insulated wire in between the two ends of the plate resistor.

Is it possible this would work in series instead of in parallel? So it would go: resistor=>smoothing cap=>pin 1. Thanks!

[sluckey]

Is it possible this would work in series instead of in parallel? So it would go: resistor=>smoothing cap=>pin 1. Thanks!

no

[marshallguy]

I've tried to do this a few times but there's not enough space in my amp to try it, a big cap and an insulated wire in between the two ends of the plate resistor.

Is it possible this would work in series instead of in parallel? So it would go: resistor=>smoothing cap=>pin 1. Thanks!

I’ll translate “no” like the white cop did on Sanford and Son when he would try to explain to Fred and Lamont something and they’d look at the black cop and he’d explain it in a few works they all could understand…….like “no”…LMAO😆

Ok, now we got that humor out of the way, if you put the cap in series, you’d be blocking the DC from the B+ P.S. to the plate of the tube. No voltage = no current = no sound = heavy drinking = more questions = no booty, etc..
In parallel, the higher audio bypasses the plate resistor unrestricted to filter cap audio ground.
That’s what they want to do, pass very high audio harmonics to ground. Smooth it out.



“Failure is not an option”

[sluckey]

I think you are talking to a troll who disappeared shortly after making that post.

[marshallguy]

I posted this schematic over at AG, but thought it would be helpful here also.

Mesa uses a similar arrangement on the Lonestar amps.  Mesa's implementation differs in which "side" of the PI plates has the cap tacked on.  Theirs is opposite of yours.

*120pf across the 100k PI plate.

Randy Smith has changed the P.I. and other parts so many different ways over the years, he should be asking questions too! 

If you like to work on amps, try his Rectroverb 50 amp. It will keep an amp tech busy…