Xits X10 = liverpool half power?

[azatplayer]

Paolo, when i did this clone ages ago, the tone controls stumped me too. And it had more gain than was usable. The bass control worked oddly to say the least, and the treble control was more of a gain control than a treble control.
They say thats what it does tho, increases gain and treble response, but i found it to get thinner rather than brighter.
What bothered me was that the tonestack feeds the third stage directly, so its gotta act as a gain control of sorts, but i could never get the bass and treble to behave nicely till i placed a .02 cap in series from the treble wiper to the grid on the third stage.
Actually edit that, i think from memory I had a 100K from the wiper into a .02 cap into the grid.
I didnt play with that circuit for long, found it frustrating, AND i had ordered one by that time which ive since abandoned. I ended up building a really nice 15 watter with very voxish vibe that is doing all those tones for me with the help of VVR.
Ive said to a few guys here, the tonestack is not kosha. Theres something we cant see going on. I also think i have seen a kinda fat looking bit of heatshrink coming from the grid of that tube in some pics, like theres a cap in series that cant be seen, or a resistor.
I know a 100K on the wiper of a JCM800 style master, helps that control work much much better.

[pamaz67]

nice discovery.
Looking at the photos i think it's quite difficult to be certain that the preamp plate and cathode couples are

220K - 1k5
100k - 2k7

So I've swapped the cathode caps, that means lower gain and less bass in the first stages.

so now it's
220k - 2k7
100k - 1k5

and I like it.
secondly. I put a 1uF cathode caps instead of the 2,2 uf. Even better.

The gain starts to be manageable.

Now I'll try to put again the 1Mohm pots in the tone stack and see what happens.


an update: The 1Mohm tone pot in this arrangement are the more correct.
I removed completely the 1 uf bypass cap at cathode nr2. Better either the gain regulation than the tone........

[Masco]

AND i had ordered one by that time which ive since abandoned.

So, you've cancelled your X10 order? I guess we'll never know the real recepie.

[surfsup]

paolo, thanks for trying the 1uf caps. Interesting you have only one cap now.

Also, regarding Ra/Rk combinations, I believe this is correct:
220K - 1k5
100k - 2k7
In the pictures, the 220k connects to the relcap to bright switch, and also to the tube (gray lead), with grid input (big white lead under board), and white cathode lead. White cathode is connected to 1.5k so it seems to be 220k/1.5k

The 100k to gray tube lead to the black insulated lead with the black cathode lead next to it. The black is to 2k7. so 100k/2k7

azat, you are right, there is a photo I have of an x10 head with a resistor on the grid of tube two. Last photo, but it does not seem to be present in the Sadie. But I don't see a cap in any of the photos.

[img:400:531]http://chicagocadcam.com/ChrisHahn/schem/sadie1.jpg[/img]
[img:400:531]http://chicagocadcam.com/ChrisHahn/schem/sadie2.jpg[/img]
[img:400:600]http://chicagocadcam.com/ChrisHahn/schem/head1.jpg[/img]

[pamaz67]

surfusp, I think it's correct as you say.
But the sound with less gain and a little more sparkle is much more close to what we are searching for .........
BTW.

Let's try to make a little bit of order.

1) the amp, built as per schematic v2.1 unfortunately is far form the clips. It has way too much bass, and is muddy. Is way too gainy. The tone stack is not working in a decent way
2) azat suggested to "isolate " the tone stack from the 3rd stage via a resistor and/or a cap. I've verified it both with resistors (22k and 100k ) , than with cap and reistor in series. Yes it works in some ways but it's far from being optimal.
3) To take out a little muddiness, it's possible to reduce the bypass caps on the 1st and 2nd stage
4) working on the tone pot values is not solving any problem of gain/tone Therefore is not the way to go by now
5) 12Au7 is mandatory.
6) The supply voltage at the pi is IMHO too low.

Any ideas?

[azatplayer]

You can see in that last pic of the head version, (its an early design, before he dropped the divider and grid leak resister), there is a resistor heatshrinked into the shielded cable return from the treble wiper.
Could be just a grid resister of maybe 10k or so, or something more to isolate the tone circuit.
Yes Paolo, it was far from optimum. There is still some mystery to be unfolded for sure!

[pamaz67]

You can see in that last pic of the head version, (its an early design, before he dropped the divider and grid leak resister), there is a resistor heatshrinked into the shielded cable return from the treble wiper.
Could be just a grid resister of maybe 10k or so, or something more to isolate the tone circuit.
Yes Paolo, it was far from optimum. There is still some mystery to be unfolded for sure!

Yes AZAT I've seen it.

And I was speculating what a reasonable value could be that hidden res.
I'm assuming that the 56k grid leak removed and the 12ax7 were "traveling" together

removing the 56k partition and using a 12AU7 is something that tends to leave the voltage at grid nr3 in the same range, I think.

furthermore, the 56k resistor was the dominant load for the tone stack, thus the huge variation in volume) when you insert/remove it. ( 12db roughly as per the tonestack calculator)

Thinking about this, a 10-22k is worthless because is not acting as an efficient voltage divider. Something in the range 50-200k is much more plausible to cut the voltage at the 3rd stage.
An higher value seems to be not reasonable because having the 3rd stage a 10kohm cathode resistor, it can get some decent voltage at its grid before going in grid conduction.
I mean this is a pure speculation, but is the only plausible way that come to my mind, so to define a value for the hidden grid resistor.
Therefore 100k is a reasonable value I can leave in that position

[surfsup]

Paolo, regarding Rk, doing the math for V1a:

Ck=5uF:
Ri=180k (max volume 220k Ra parallel with 1M = ~180k)
R=2k7 fboost=14Hz gain=61
R=1k5 fboost=24Hz gain=61

Ck=2uF:
Ri=153k (220k with pot at 1/2 ~500k)
R=2k7 35Hz, 59
R=1k5 60Hz, 59

Ck=1uF:
Ri=117k (220k with pot at 1/4 ~250k)
R=2k7 69Hz, 55
R=1k5 120Hz, 55

So it seems the Rks would have very little audible effect until the Ck value was reduced to about 1uF. 2uF shows a swing from 35Hz to 60Hz which is barely audible unless the amp is cranked. So 1.0uF seems to be where your testing should start?

Also, you are now using 220k/2.7k and 100k/1.5k, the reverse combination of the Xits Ra/Rk. So you are cutting a lower frequency at V1a, then a higher frequency at V1b where the Xits X10 design is doing the opposite (Ck=1uF):

Paolo:
220k/2.7k f-boost=35Hz
100k/1.5k f-boost=62Hz (and I think you even removed the 1.0uF here which would reduce 2nd harmonic distortion?)

Xits:
220k/1.5k f-boost=60Hz
100k/2.7k f-boost=37Hz

So it seems the Xits X10 is boosting the higher frequency first, boosting less low frequencies so those will not be further amplified by V1b, then boosting them at V1b instead so overall through the amp chain the frequency range of 35-60 is being boosted once, not twice. Am I correct in mentioning that?

As I have no idea how to calculate the overall resistance of the tone stack acting as Ri for V1b, for the 3rd gain stage (V2a) removing the 56k is removing a parallel resistor to the tone stack which has a much higher resistance than 56k I think, and removing the 56k would effectively increase the separation of stages 2 and 3? This one is just a guess...

[pamaz67]

When I was trying to summarize the situation I've dealt with,
I was referring to this configuration
v1
220k- 1k5 - 1uF
100k- 2k7- 1uF

Tone stack and any other spec as in V2.1 ( 1Mohm pots) aside from the power amp that is 4xel84.
Any cathode cap larger than 1uf cause low freq. muddiness in MY amp.

As I told there are different issues on the floor.
The main is the not correct way the tons stack works.
Now I'm trying to understand what are the possible issues that can arise in that area.
I.E how the tonestack is loaded in an AC30?........

[pamaz67]

How the tonestack is loaded in an AC 30?
Azat you are right.
220k+47nF in series toward the grid of the following tube.
http://www.voxamps.com/downloads/circuits/ac301986.jpg
Now I'll do it and see what happens.

As a side note on the same drawing: if the x10 is similar to an AC30 the correct thing is 220-2k7 and 100k - 1k5 . Look on the same schemo.

[pamaz67]

probably we can say BINGO!
no need for the capacitor, but if we put 220k from the wiper of the treble pot, to the grid of the other tube, we have a tone stack that works.
And (listen to the news) you may want to place 25uF on both cathodes of the first stages.
I still prefer a lower value , but the muddiness now is gone.
The gain is high, but with single coils you can dial also some clean tones.
With humbuckers, better to go wild and set a nice overdrive tone and clean with the volume of the guitar.
Probably we are dead on.

[surfsup]

Paolo, that's great I was just about to post that I saw a 220k on the X4 in a photo and wondered if that would be something to try. AWESOME!

I have a couple 220k 1W already. I PM'ed you if you tried the 12AT7 as a PI tube. Would be interested in what you think. Could you post a couple gutshots and a voltage chart when you have a chance? i would very much appreciate it.

[pamaz67]

Pls. Colossal, Could I ask you to update the schematic indicating that resistor with a tentative value of 220k?
That's mandatory for having a working amp.

I have tried also 470k and the behaviour is quite similar to 220k . Not a great difference.

Tomorrow I will test also a tiny mod that with an additional switch could be useful to tame a little the gain of the beast, ( or to leave it original) .....

[surfsup]

Still using 1M pots for everything?

[pamaz67]

yes 1Mohm everywhere. just like the vox thing that i posted.
Finally, the x10 is a "barebone 1986 ac30" with an additional gain stage and with half the power tubes and a lar mar.