Tech 21 XXL done and .....wooah momma !!

[MartyMart]

http://www.diystompboxes.com/smfforum/index.php?topic=29094.0

Thanks stm, problem solved !
OK, I'm waiting on some TLC2262's which I'll try asap but here's what I did:
Ignored all the switching/buffering totally, input from the 27n/10k to ic pin 3 with the
1M bias etc intact ( ignored the second 27n cap )
IC ( for now ) TL072CP
Ignored the small "ohm" resistors across the power line, used a 1N4002 rev protection diode and
220uf  10k/10k/100uf for VB.
Warp pot 100k linear, helps to set bias more acurately I think.. ( subtle changes here as noted above )
Gain 100k log   and used the suggested 10k for R8, for a lower min gain setting
Tone 100k lin ( all I had )
From the tone section, direct to a vol rec setup, 390k/100k base, 10k to collector and
3k3 emitter to ground, using an MPSA18
Out cap 1uf poly and a 47k vol pot ( more than enough vol here !! )
Took about three hours, including a small de-bug ( didn't connect 9v to the IC  ..!! )

Result is VERY good indeed, I borrowed a real XXL and I think it sounds better, probably due
to all the switching being removed, etc ... as I said TLC2262 will be tried as well.
Considering this is a "NO DIODES" crunch box, it's a great sounding one ...

oh and it fits on a piece of vero 10 strips X 28 holes, so "smal"l but its 1590BB due to all the
pots !
Happy person
MM.

[troubledtom]

cooooooool, are you gonna post the layout
           peace,
               - tom

[aron]

Ahhh the Insanity Box is better :-)

[MartyMart]

Ahhh the Insanity Box is better :-)

You wouldn't be "biased" there would you aron ?

cooooooool, are you gonna post the layout
           peace,
               - tom

Yeah sure, I'll draw it up in the morning

[lowstar]

cool, i´d love to see the layout too !!
cheers,
lowstar

[aron]

Hehehe, the Insanity Box was designed because I did have an XXL in my studio.

The good old days.

[stm]

Congrats Marty! 

The change in R8 shifted down BOTH min and max gain by 3 times. This may also lessen the maximum assymetry effect, which is achieved when the first opamp clips.

Instead you may try replacing R5 with 1k and R6 with 330k. Now you have same max gain with a minimum gain of one third the original, and if you use a log pot for gain now you have a 100:1 gain change ratio with a very smooth and even gain transition throughout full rotation (which is what I like for log pots, i.e. the series resistor is one hundredth of the pot's value). In addition, the use of a smaller resistor in the feedback loop of the second opamp might reduce background hiss a tad.

Also, it is possible that the gain recovery stage you added (with gain 3x) will alter the final sound if it reaches clipping, which will do at high LEVEL settings. In the original they only use gain 1x stage for this purpose (this is why I suggested using the tranny as an emitter follower).

Regards.

[MartyMart]

Thanks stm, that makes more sense !
I'll make changess in the morning and report back ...
So for the vol rec, your suggestion of a 4k7 from emitter gives a gain of 1 .. yes ?
Should I also make this a 2N5088, or stay with the MPSA18 for quality ?
For the "warp" control, I've used 2 X 33k's to lugs 3 and 1 of the "100k linear" pot, I'm
thinking that will make "centre" even for bias, hope this is correct
original was 22k and a 47k, allowing for the 100k log pot's rotation ...

Regards,
Marty.

[stm]

Marty,

The 4k7 resistor goes from the emitter to ground; you place a short (no resistor or zero ohms) from the collector to Vcc (9V); you take the output from the emitter instead of the collector. This is an emitter follower, meaning it has almost unity gain (in fact it may be in the range 0.99 to 0.999 depending on the transistor gain). The 4k7 resistor on the emitter just establishes how "strong" your buffer is. That's a reasonable value. However you can move between 1k (at the expense of more current consumption) and perhaps as high as 10k (at the expense of a higher output impedance). Within this range the gain will be essentially unity.

Regarding the transistor type, the 2N5088 is more than adequate for this purpose. Try them both and if you can hear a difference it's either because you have a privileged ear or your listening tests are biased towards a particular transistor.

On the warp or xxl factor, the idea of having DIFFERENT resistors on each side of pot is not so bad. Someone pointed in a previous thread on the XXL that both extremes produced "typical sought after" amounts of asymmetry, one slightly assymetrical and one rather fuzzy, so it gave you the chance of having preset values.  If you don't want to be this way you can make sure one end of the pot is essentially at the center and the other is biased towards one side, meaning your control will go from symmetric to asymmetric. In order to achieve this, and assuming you have a 100k pot, you can place a 150k resistor on one side and a 47k or 51k resistor on the other side. In this way, with the pot on one extreme you have 150k and 100k+47k which is basically centered (Vcc/2). On the other extreme you will have 150k+100k and 47k, which would produce a bias of 1/6th or 5/6th of Vcc.

The only drawback I see is that pot resistance tolerance is quite loose, and you may easily find a nominal 100k pot measuring anything between 80k and 120k, so eventually you may need to adjust the surrounding resistors.  BTW, I think maybe this is the real reason behind the odd selection of resistors made by Tech 21 in the XXL, make sure that regardless of the pot tolerance you may always find the 50% bias point. Nevertheless since this is DIY and not high volume production you can afford doing some matching in this case.

Finally, this pot should be LINEAR for best results.

[MartyMart]

stm .... you ARE a god !!
I'll incorporate those wonderful improvements on my layout drawing tomorrow

I LOVE the sound of this thing.... a very "original" and wierd tone !

MM.

[Steben]

Again, a simple circuit isn't it (if you discard the switching). It's all about those 2262's...
Never thought assymetry could make such a change. It really sounds like a smooth saturated fuzz doesn't it?
Would it make any difference using the other half of the 2262 instead of the TL072 in the gain line (and using true bypass)?

[MartyMart]

I'll be trying a 2262 in there tomorrow, though I'm very happy with the tone right now
we'll see ......
As it's an op-amp clipper and "diode-less" it's probably close to "Hot cake" land ....
similar idea, but with a single TL071

MM.