Ibanez TSA30 Schematic and PCB Layout

[Mprall00]

You could replace R72 with 560k and change your new gain pot with 250k.
Also replace R4 and R5 with 100k,, that will back off the gain.
Is R13 (56k) still in circuit?

You can alter the tone stack might help back off excess treble.
Make C23 100nF and C22 22nF might work,, not sure how well but suck it and see.
Phil.

Cool thanks.  I will try it!  r13 is still there.  I have a 510k metal film on hand to replace r72 with and a 250ka pot.  So I will probably try that as well as the 2 100k changes for r4 and r5.  i already replaced those with metal film in my quest to get the noise floor down but will sub in the 100k and see how that impacts it too.  That 50k difference in your recommendation at r72 won't impact much will it?  Slightly more gain?

I already changed the tone stack to more standard fender cap values.  C23 to 100n and C24 to 250p.  Left the mid cap alone at 47n for now.  The tone is good.  A tiny bit bright but treble knob is just running lower than normal and sounds good clean.

[Mprall00]

I went with 510k.  That with the 100k for r4 and r5 and the 250k gain pot changed the eq of the amp quite a bit.  I think i am going to up the gain pot to 500k now and thats it!  Works great with pedals.  If anything now it is too bassy and the the treble is way more tame.  I may play with values a little more but I think its a pretty happy medium now.

Here is another thing I plan on playing with - a switch to bypass the u3 solid state front end instead of using 2 inputs with a switch to select between them.  But... that will leave the buffers in place from the TS and boost so that may change the tone of having a input straight in to r13 to V1...

I'll take ideas if you have them.   

[phatt]

If R13 is (56k) then it's likely done to tame the amp down as the two SS dirt stages would be too much for the Amp section.
Normally R13 would be at least 220k,,more common value is 1Meg.

With 1meg at R13 will open the amp up huge but you will then have to re eq the whole amp circuit.

You can play with a few sections;
Remove C51 for a bit more top end, if you still want more top remove C50 as well but watch out for feedback at high volume.

If R72 is 500k you can add a treble bleed around it  ~470pF is common.
(just go find all the classic valve amps for clues)

You can fiddle with the values of R18 & R25,, dig up some bassman/Marshall schematics for ideas. That changes how the PI distorts amoung other things.

If too much bass make C10 smaller ,try 1n5.

Other stuff;
You don't need metal film for plate resistors normal carbon is fine.
you use metal film for signal path not DC stuff.
As to your ideas of the SS crap in front,, it's anyones guess how it will help,, frankly the whole thing is a bit of a miss mash.
I'd focus on getting the Valve Amp section to work well then down the track you might be able to tweak the TS part.
I'd bypass the the gain section around u3A as it's likely too much,, hence they had to turn down V1 (R13 56k)

With a pair of 6L6 running from 560VDC would be way over 30Watts so it should be loud if not then they don't know how to build Valve circuits.
Phil.

[Mprall00]

Good advice.  Question:  I managed to bypass the u3a but I bypassed all of u3 so lost the boost too.  I pulled the leg of r33 and went to cn3a from that spot.  Can I keep the boost in place by starting somewhere else?  Not seeing where that would be.

[phatt]

My advice remains the same.
You need to get the amp to work FIRST.

To do that;
Lift the R62 side of c30 use a shielded cable grounded at one end and join up with pin 2 of CN3a.

Lift one end of C27 to remove all the SS stuff.
Make R13 1Meg.
Now you will have an all Valve circuit with a normal input.
Work with ideas already mentioned until you get the Valve circuit to work and sound good,,,,
Then try ideas with the SS part,, IF it needs it.
Phil.

[Mprall00]

Interesting... so i bypassed all the SS by literally removing it from the chassis (unplugged the connectors for CN1a, CN2A, CN3b, and CN7a) and bolting a new input with a 1meg from input to ground just acting as a pull down resistor to the existing input hole.  Then ran that input and ground to CN3a. 

Sounds like we accomplished the same thing because i then had the input straight in to CN3b which is r13 (which I had changed it to 68k because that was what i found in other fender schematics).

But I am missing your 1meg in line before v1 right?  I don't see that in other fender schematics.  Seems very high.  Is it serving a purpose there?  Doesn't seem like it would do anything to the tone right?  68k in series with 1meg before v1...

Thanks for your help!  Learning a lot.

It sounds good without the 1meg in series, so i think I have the tube amp part solid.  Just wondering how the 1meg would effect it.

[phatt]

I think you missed something,, The 1 Meg Replaces R13,, leave R62 in place.

Here is what it should look like as a schematic,, there are 1,000 of schematics on the net and almost as much info. I've numbered the resistors so they match your schematic.
Google, Valve wizard,, Tone Lizard,, Rob Robinette,, and many other who have good info.
Phil.

[Mprall00]

I knew i did something wrong!  Thanks.  Totally misread and thought R13 was in series in to v1 and didn't notice it was going to ground!  Stupid... i think I have it now.

[Mprall00]

Thanks for all the help.

So the tube amp portion now sounds excellent and plays nicely with pedals and is much quieter now.  1meg is in place for r13.  Switched r18 to 470r.  left r25 at 15k since that is between a bassman and deluxe reverb.  May play with it someday but sounds good now.

Still pretty bass heavy.  Could play with c10 (or c54?) coupling cap but don't have any on hand rated at high voltage.  Just running bass control around 30% and it is good clean but as the new gain is turned up it distorts pretty bassy.

Now thinking about how to work the SS portion back in to the circuit on a switch.  Have an spdt on-on I am going to use maybe at the point you mentioned with r62 going to middle lug and one side going straight to cn3a and the other to C30.

[mikemagyar]

Hey Mprall00,

How did you go about wiring up the 500k pot to replace r17? Did you just wire it as a variable resistor?

I've followed this thread and modded my tsa30 to have the all-tube signal path I'd hoped the amp would have when I bought it! However, without some means of attenuating the signal before the factory volume knob(i.e. the r17 pot mod), the signal is slightly distorted.

What's the advantage of replacing r17 with a pot vs replacing r13 with a pot? Could both r13 and r17 be replaced with pots to control signal gain at both stages?

Thanks in advance for any and all knowledge shared!

[Mprall00]

I just wired it as a variable resistor.  Actually just got a 1meg pot because the 500k pot at R17 was not breaking up much when cranked and even though i almost always play it clean as a pedal platform I thought having the option of more preamp gain would be nice.

I guess you could replace r13 with a pot as well to control v1 too.  Let me know if you try it!

After tweaking a lot of resistor values and replacing the tonestack, the SS front end actually doesn't sound bad either.  I got the solid state section to be switchable with the stock input jack but it doesn't sound good.  When the SS is bypassed it sounds like the impedance is way off - like something at the input needs that buffer from the SS sections.  I even tried adding a 1 meg pull down on the stock input jack...  It just sounds really flubby with no presence when the SS is bypassed but sounds perfect when i removed the whole daughter board and added my own input jack straight to CN3b.

SO... i am about to drill a hole next to the existing input for a second input direct to CN3b and a hole for the gain pot above and between the gain and bass knobs.  That way I can just plug in straight to the tube section with the new input or plug in to the TS>Boost>SS front end if i want to use it.

Side note: replacing the signal path resistors with metal film and replacing the coupling caps and tone stack made this amp a lot less noisy.  Not sure if one of those thing did more for the noise floor than any another - i just know the whole change helped.  i got mine very old and very used, so it could have been noisy due to age, but this really helped as did a proper bias.

[phatt]

The problem with the SS section is that the gain section is on all the time.
Which means those Leds (D11-D15) will add some clipping (Dirt) to the signal even when boost is OFF.

You could try removing the gain section.
Break connection at Q7 pin1, also the top of C27.
That removes the drive section.

Now just join pin2 at CN2a and then join that to pin2 at CN3a.
That gives you the always on input buffer as well as the switchable TS circuit and FX loop is still working.

Remember I don't know this amp so it's anyone's guess as to how well this would work,, personally I'd remove (and bin) the SS section and use pedals.
Phil.

[Mprall00]

I agree... But if i toss it then i have 5 open holes in the front.  And if I keep it and break the connection as you outlined then i still have to drill a hole for the new gain pot for v2.  And if I am drilling anyways I am going to just solve it with a second input.  So my new input will bypass all of it - buffers, ts, SS, everything - and go straight to the CN3a.  And the stock input will still be there and use the stock daughter board as is.

Truth is, once we changed all that stuff (i changed a lot with your help but the biggest impact was the 12v regulator, R4, R5, R13, R72, R17 as a pot, and the tone stack caps), the SS front end actually doesnt have a bad gain tone.  So therefore I think I will drill one extra hole and have both options, even if I don't really ever use the stock option... Just turn the SS gain all the way down, plug in to the new input, and it will be like it isn't there.

Once done I will try to remember to take pictures and write up an easier to follow list of changes in a new post.  These can be fun to learn with since they are relatively cheap for 30 watts.  I got my TSA30 combo for just under $200 used...

[phatt]

Yep a fair point.
As I don't have that Amp I'm only guessing.

One thing I should add is that the Schematic shows no Grid Stopper resistors on the power tubes and although you won't notice much at low volumes when the Amp is working at high volumes it will most likely spit,, sometimes called power stage choking.
I'd suggest at least 5k and right at the pin if possible. you may have to cut tracks to do it but worth it if you play with a lot of heavy loud distortion.
This very problem drove me nuts on some of my early valve amp builds years back as I only had 2k2 resistors. sounds gawd darn awful.
I replaced them with 10k and bingo no more harsh blocking distortion. 

Explained well here if interested;
http://ozvalveamps.org/choking.htm
Phil.   

[Mprall00]

Good advice!  Didn't notice that.  I bet it was never a problem with the original design because it was never getting loud enough to get the power amp cooking anyways.  it was so quiet stock...

I doubt I will get it cranked that loud anyways but in case I do...

Would that mean essentially cutting the trace between R14 and pin 5 of v4 and R10 and pin 5 of v5 then adding something between 1.5k and 10k where i cut the trace?  Is there any downside?  Nothing mentioned in what I am reading... did ibanez just design it without to save $.05?