Ibanez TSA30 Schematic and PCB Layout

[phatt]

Yes highly likely you will have to cut tracks.
As I don't have a track layout I have no idea of how easy that would be.
I've found with pcb's you often have to get inventive with solving issues. I find a week may pass sometimes before an idea comes to mind, other times it comes to me straight up.
If the tracks are wide enough you may be able to cut and drill mounting holes.
but a lot of those pcb's have very narrow tracks which makes it hard to find a spot.
If tracks are too narrow I drill beside the track and solder the pig tail along the track.

No down side to grid stop resistors that I've read about.
I think you will find it's Mandatory on nearly all good designs.

As to why they omitted those resistors,, anyone's guess.

Be aware a lot of these amps are just whipped up in a mad frenzy with little or none R&D.
The whole amp is likely out sourced to a far off country,, read cheap labor and no QC

There was a HiWatt Amp question on here long ago that was so badly designed it was unusable because of noise due to bad PCB layout.
turns out HiWatt had little to do with that Amp and hence had no records.

AFAIK, Marshall still make amps that use EL34 tubes and run them at elevated voltages,, which is no problem for the plate of EL34 but the screens are then way past their max voltage limit. For most brands of EL34 the screen limit is 420volts.
They don't last very long with the screen at 470volts.
Do they care? I doubt it.
The RCA manual states quite clearly that AX7 will last for 10,000Hr's and power valves about half that.
The Valves in a well designed amp would/should last for at least 10 Years. (with heavy use)
I've witnessed Amps that are 30 plus years old and still using original Valves.
Today they just slap it together and know that the valves will last at least a year.

A chap who came to me years back had asked the local music shop,, How often do you replace the valves?
The shop owner replied,, oh every year or so.

He was so glad he came to me for some advice.

Phil.

[Mprall00]

I'll have to look tonight but i feel like C13 and C5 being so close to pin 5 means i could almost cut the trace and then just solder the resistor to the existing hole that the leg of C5 is already in then directly to pin 5 without modifying the board at all.  Any problem not using the board at all here?  More noise?

[Mprall00]

Another question while I am at it - V1 and V2 don't have a cathode bypass cap.  Worth adding one to pin 3 of both?  Or at least V1?  would raise gain... and depending on value, on specific frequencies?

[Mprall00]

Coming back with another question - can anyone tell me how the 2 triodes of v1 and v2 are working?  Staring at it realizing they aren't parallel.  And they aren't in series... are they?  is the signal coming from the cathode of both v1 and v2?

[phatt]

You have 2 Cathode followers, one at V1 and another at V2.
Implementing a Cathode follower at V1 is likely just a hair brain idea by someone in design who reads too much internet gossip. probably no great advantage over a single triode.

Obviously you did not google Valve wizard.
It's all there for you to learn. Hint!!!

http://www.valvewizard.co.uk/dccf.html

Main Index;
http://www.valvewizard.co.uk/index.html
Phil.

Regards bypass caps at V1&V2,, yes try some values and see what happens.
Not a lot of advantage at V1 but at V2 will pull more tone change.

As it's an oddball circuit (with the SS preamp)
It's anyone's guess how much advantage adding Cathode caps will help?
Phil.

[Mprall00]

Haha. Late night dumb question. I’ve even added a cathode follower to a blues jr. Think I was thrown off because I’ve never seen one on v1 as I was looking up schematics... and it was late. Thanks for the response to the dumb question!

[Mprall00]

Any advice on adding NFB so I can add a presence control?  Tried a couple things and actually deleted a post from yesterday that didn't work.  I tried deluxe reverb style later and got a loud screeching noise.  No damage done but definitely didnt work.

For that I tried lifting the grounded side of R15 and C16 and connected them. Connected 820r resistor from the speaker output to the lifted sides of R15 and C16. Put one end of a 47r resistor to ground and the other end to the connection of the other 3.

The end goal was to add more like a bassman style with a presence control.  Just tried without to start and didnt work at all.

Edit:  Flipped primary because that sounded like positive feedback - same problem.  Hmm....  Definitely seems like I am getting positive feedback either way...  It did firm up bass and treble seems to sustain better.  It is also in phase with my blues junior when run in stereo now.  So all wins, but still NFB does work.

Edit 2: Tried the NFB of an AB165 bassman (47k in series with .1uf fed back into the phase inverter after C10).  It looks like weird compared to most NFB set ups that i tried originally.  It worked this time since it is actually negative feedback, but it reduced signal way to much.  Not sure why the much more common place NFB from other fender models squealed like positive feedback.  Research continues...

[phatt]

well you had me wondering as I noted a posting and then it was not there.
Not to worry, in one sentence;
If you want lots of distortion then don't bother with NFB.

The NFB of AB165 bassman is not going to work without major mods. you need those 220k resistors from the plates of power tubes. It was not a good design.
(look up Rob Robinette site for lots of clues on what works and what is a waste of time)

I'm not a big fan of presence systems as you need NFB and that will run the amp cleaner so if you want lots of OD then you have to run the amp at much higher level.
If that is what you want then go for it.
If you just want more treble then lower the coupling cap in front of the PI stage.

Look at some fender schematics for clues,, a lot of the AB763 circuits have very small value caps at the PI input.

Those big fender twins are so loud due to the 1nF cap at PI input, Change that to 22nF and they would just fart out at high volume.
Remove the NFB and they would not be so clean at high volume.

Regards to Bassman,,, Which one?

Old valve amp schematics might look simple but you need to build a few to realise that a couple of changed values can make a huge difference.

At the heart of most of them is the Power supply resistor values, they are often very different and that has a massive impact on the sonic result.

As example;
Your Amp circuit has R3 (22k/2Watt) now go look at most Bassmans and you will see they are 1k to 5k.
That alone makes a big difference as to how early the PI distorts.
Your circuit is closer to Plexi Values which means the PI OD's early.

Most Bassman's use AT7 while nearly all marshall's use AX7. Again that changes how the amp responds.

I could go on here as the options are endless but maybe you can just print out a dozen Schematics and look closely at the Power rail values.
Meantime read up Robs pages.
https://robrobinette.com/Amp_Stuff.htm

At 750 pages you might be there for a long time

HTH, Phil.

[lighthunt]

First of all, thanks for the schematics! ;-)

Gentlemen, did anyone tried to mod/add real FX loop to tsa-30? IMO it's the biggest weakness of this amp from player point of view. (I'd like to gently overdrive the signal with TS  and push the tube with higher volume/gain to get sound I like, which works well on it's own but adding reverb in the front or into the "fake fx loop" ruins it).

So I am considering to add fx loop in between second preamp tube (V2) and the main volume pot. This should be easy mod since there is shielded cable already, but ... will such simplistic/passive approach work well or do I need to add any buffer there (transistor or opamp)?  I've studied other amp schematics (e.g Peavey classic 30) and they have transistor on the return. Is it necessary? Why? (impedance matching or so ... ,??)

I have some, but still limited knowledge, so I really appreciate your recommendation. Thanks!!!

[phatt]

The circuit already has a loop.
Inserting it in the middle will likely have little benefit.
As to your idea the problem is not an fx loop it is the design of the circuit.
i.e. a ss front end designed badly and little you can do without major mods.

Now note I've never been close enough to smell one of there amps but I would not be interested after seeing the schematic.
It is possible some clever chap has found a way around it but I doubt it would be simple.
There is too much happening at the front end so as you note inserting a reverb FX in the loop does not work well.
If you are hell bent that it will make a difference do it at R72.
As I have already mentioned,, I would bypass the whole front end and connect the input to V1A and change R13 to at least 470k. THEN the circuit will (Likely) be more pedal friendly.
Better mind might have more knowledge,, Phil

[lighthunt]

Hi Phill, thanks for response!

> "The circuit already has a loop."
yeah, but it's between the tubescreamer and tube preamp. so pedals work well if I keep tube preamp clean or almost, but not great dirty (for reverb etc).

the SS frontend is the tubescreamer, nothing wrong with it and can by bypassed by pluging guitar into this fake FX loop return if you don't like it (well almost, I think there is still a clean buffer before tube, no truebypass switch).

To my ears, sound is great, lot's of tonal possibilities, just missing that real FX loop between preamp and power stage. not good for live playing if you want to switch between clean and (amp) tube distortion, cause you can never get the volume right for both it's one channel combo with swichable TS in the front. With real FX (between preamp and power) I would be able to control volume by pedal for both clean and dirty amp or proper reverb for both.

If I find time, I'd do the poor mans FX loop (https://robrobinette.com/Generic_Tube_Amp_Mods.htm#FX_Loop), should be easy mod since there is shielded cable leading to volume pot ...

Will report here later.

Cheers

[phatt]

OK but some stuff that might explain why not.
The Passive FX loop is for a 5E3 while the tone setup in your amp is the later TMB setup. The simple treble cut tone of a 5E3 is not as lossy as the BF type tone in your Amp.
AFAIK
Any loading of the volume control output will impact the effectiveness of your tone controls.
Yes it will pass signal but at the cost of signal strength  and tone compromised.
Some pedals will certainly load down the signal if inserted there.

You would likely need to use another HiZ stage so the tone does not suffer needless loss.
But As I mentioned the previous stage is ideal for driving an FX loop as it's a CF stage and will not suffer from any loading,, but as it's a pcb that maybe a hassle for you. I would Break signal after R72 and use the passive loop there.
The problem with the idea of FX loops inside a valve circuit is that you interrupt the magic of how the power supply drops all through the amp. Valve power supplies are not stiff like SS Amps and a lot of the mojo has to do with the interaction of supply rails as signal goes up. It's not just sag it's how it effects the whole path from input to output.

If you care to look at all those old 6V6 deluxe schematics you will notice that the first 2 drop resistors in the R/C supply are 10k & 10k in some but then the 5E3 it's 5k & 22k.
Just by changing those values makes a big difference to how the circuit responds, dynamics and feel change.

So if you want to get that stuff to happen then DON'T break into the circuit,, use pedals in front.
If you want reverb then you need to design the mixer part like later models of fender deluxe.
Hope that helps, Phil.

[lighthunt]

The more someone tells me don't do that, the more I want to try. Or at least experiment to find out they are right ;-) So disappointed 

No, but seriously, thanks!! for the insight, the more I look at that your answer, the more I see it replies to my original question. If I do some experiments sometimes later, I will share here.

Cheers

[bajaman]

Well, I just had two more of these abortions in my workshop - both with that nasty buzz/ hum
I tried increasing the capacitor at the input of the 10v regulator but unfortunately the buzz was still there, although reduced slightly, SO, i decided to find out what exactly is wrong with these poorly designed amplifiers and discovered some very interesting crazy design faults.
First : The grounding scheme for these amplifiers is disgusting ! The amplifier contains three boards (well four if you include the speaker ohm selector switch board) the tube amplifier board the solid state tubescreamer, boost and input gain board, and the tone control and master volume board.
The solid state board is ground connected to the chassis by way of a m3 bolt next to the input socket. There is a badly designed attempt at star grounding the tube amplifier board, including high current output tube cathodes and regulator returns, but what is truly astonishing is that the pseudo star ground on this board does not connect direct to the chassis - instead all the noisy regulator and output tube cathodes and all the ht filter capacitor grounds are connected through a thin black ground wire to the solid state input board. What makes matters even worse is that this push on connector ground link to the chassis is not even taken from the star ground point on the board - no wonder this sucker buzzes and hums like a demon. I have never seen such a poorly designed (or not) grounding scheme in over forty years of servicing tube amps - what were Ibanez thinking when they released this amplifier on to the market ?
Okay, now the fix !
This involves breaking the ground tracks on the board in several places and adding a wire from the star ground point directly to the chassis. Cut the ground returns  from C7 and C8 and connect these directly via cable to the junction of C17 and C14 (the star ground point) - next cut the ground trace from both pin 9s of V1 and V2 and connect these directly to the star ground point. Cut the track twice to isolate the middle pin of the 10 volt regulator IC and connect this pin directly to the star ground point - make sure to connect the remaining two cut tracks with a small wire connection to maintain ground connectivity to solid state board - hope that makes sense ! Remove C18 and C19 - not needed. Remove C21 and replace with C20, BUT drill a small hole on the board and attach the ground terminal to the middle pin of the regulator. Replace C20 with a larger value capacitor - preferably 2200uf at least 25v rating. Finally break the main ground trace between C6 and C16 - this isolates the high current ground from the low current ground. Now the high current ground returns are connected through the star ground point directly to the chassis and the sensitive low ground points (including the master volume control) are connected through the ground connecting wire to the solid state board and input socket chassis connection. I also removed the black ground return wire from the pilot light led and connected this directly to the chassis - never like dirty led ground currents anywhere near the input stages - this may not be necessary but worth doing i feel.
NOW - enjoy buzz and hum free operation at last !
Sorry - this will not reduce the aforementioned hiss (yet another bad design feature)  some of Phatt's suggestions will fix this - i would suggest moving VR8 the master volume to replace R17 and connecting the wiper of the treble control VR6 directly to C10 - that should reduce the hiss and work far better  - UPDATE: I did this yesterday and it made a huge difference - much lower hiss !
cheers
bajaman

[Diydidi]

Hey Bajaman!
Thank you for your detailed explanation of the mods you do.
I will give it a shot this coming week. Customer dropped off his amp and its buzzing like crazy.
Someone already installed the 2200uf cap. I tested the ESR of all the caps and they are very low indicating good caps generally.
Any idea as to why some of these amps are buzzing and others not?
Also as to the heater voltage for V1 and V2: Does it matter that they are at ±10V and not the nominal 12V as indicated on most datasheets fro 12AX7's?
My first post here. Thank you!!