Ibanez TSA15H poweramp in

[anotherjim]

I think...
Tone stack source = 100k (driving tube anode resistor).
Tone stack load = 1M (volume pot).

[thermionix]

The plan is to move the high/mid/low controls where TS controls are now. What if I connect some pots (eg voume) with wires instead of directly to pcb? Can it cause any sort of problems?

Just keep the wires as short as possible, and push them up against the chassis.

[swever]

So I kept my amp removed from the cabinet for a few weeks now - with an extra "middle" pot hanging on wires. This gave me an opportunity to carefully examine the tubes while the amp is on. At some point I noticed that one of the 6v6's started looking a bit darker (burnt/oxidized) inside, and was glowing a bit dimmer than the other one. I tried tapping it a bit with my fingernail and found it was very microphonic. I did not really mind that because the amp worked great.

Yesterday I turned it on again to find that that tube does not light up at all. Tried tapping it - no sound. It's dead, apparently. Surprisingly the amp still works. And I don't think I hear any difference in sound.

I doubt I can return just the tube(s) on a warranty. Returning the amp is out of question obviouly (and shipping would have been too expensive even if I did not mess with it and could return it).

Now I have a shipload of questions.

1) Why did this happen? Is it just a bad tube (or broken during shipping)? Is there any chance that something else is wrong with the amp and it's going to kill the next tube?

2) How come the amp still works?

3) How do I go about replacement? Do I need a matched pair? (What does that mean exactly - how are they matched? What happens if I use two tubes that are not matched?) AND... how do I bias? (And also what exactly does bias mean/do?)

4) Just out of  curiosity - what happens if I remove the faulty tube and turn on the amp? If I remove both power tubes?



PS I know I should probably read a book or two about tube amps - I'd be grateful for any pointers.

[thermionix]

If the tube quit lighting up it has an open filament.  99% of the time it's just a tube failure, not caused by anything else.  I think you said your amp came with Chinese tubes, which are generally not that great.  The most rugged 6V6 made today is the JJ.  The Russian ones are not any better than the Chinese.  You can still get NOS American 6V6s for somewhat reasonable prices, compared to 6L6 or EL34 types.  If interested, search for "JAN" 6V6 which stands for "Joint Army Navy" in other words military surplus.  Apparently the US military had millions of 6V6s made but never used.

The amp still works because the other 6V6 is still working.  It is running in Class A at really low volume.  If you were to turn the amp up loud it would sound horrible once it crossed into Class AB territory.  I recommend that you don't continue running your amp until you get a new pair of tubes.

Yes, you need a matched pair.  In the simplest form that means matched for idle current.  Most people consider +/- 5mA to be matched.  The bias will need to be set for the new tubes.  This is done with the amp running, and means connecting meter leads to potentially deadly voltages.  It's not that hard, but should be done by someone who knows what they're doing.

If you turn the amp on with no power tubes, you will have no sound.  If you remove just the faulty tube, it will behave as it does now with the dead tube left in.

[swever]

Thank you again, thermionix! That's really nice of you to answer my noobish questions.

I checked the prices for JAN and other American NOS tubes and I dont think it makes sence for me to get those now. It's most likely going to be over half the price I payed for the amp. I think I shall see if I can find something locally, probably from a tech who can do the biasing.

I pulled out the faulty tube and the amp seems to work just like before. I also tried it with volume at about 3/4 and it seems to work just fine too! Then I switched to 5W mode (triode, I guess) - works too but there is some noticable hum (sounds like power line hum to me).

A few more questions if you don't mind:
1) Why do you not recommend running the amp with a faulty/only one tube? Is there a risk to damage something?
2) What happens if I put new tubes without adjusting the bias?
3) What happens if I don't use a matched pair, but two random 6v6's?
4) Can I adjust the bias "by ear"?

[thermionix]

Yes you can cause damage by running only one tube.  Most likely to the tube itself, but possibly to the output transformer.

The bias sets the operating point of the tubes.  If too hot (underbiased), tube life will be shortened.  If too cold (overbiased), there will be no damage but the tone will not be good, even horrible if you get crossover notch distortion.

If you use two random 6V6s, you might get lucky and they match well enough.  If not, which tube do you set the bias for?  Maybe you have a target of 26mA per tube (just an example) and you have one tube idling at 15mA and the other idling at 30mA...what do you do?  Some amps have individual bias adjustment for each tube, but most, like yours, do not.

Biasing by ear is possible, but not recommended.  The differences can be very subtle until you hit the extremes.  One thing some people do (again, not recommended) is crank the amp all the way up with one person playing through it, and another person adjusting the bias just low enough that the plates don't turn red.  That still involves sticking your hand into a live amp, and I see no advantage to that method unless you don't have access to a milliammeter.  Personally, I want to know the numbers.

[toykat]

I would like to clip the bright cap of my tsa15h.  Can I simply clip one leg of the C10 or must I clip both and physically remove the cap?
TIA

[PRR]

Welcome.

> Can I simply clip one leg

Yes.

[toykat]

So I snipped one leg of the C10 cap and now this is a pretty much improved amp overall and a much improved amp when used with pedals.
Does anyone know what the intended purpose of the C10 is? 
IMHO it makes no sense to have it in the circuit!

[thermionix]

It just makes the amp brighter, more so at lower volumes (dimed out it makes no difference at all).  Depending on your guitar and speakers, brighter may or may not be a good thing.  I know I wouldn't like it.  Fender Deluxe Reverb amps have a similar cap on the Vibrato/Reverb channel, I always snip those out, or at least desolder one leg on the vintage amps so it can be put back to stock if desired.

[usgmp]

I know this is a long dead thread, but in case anyone is still thinking about modding these tsa15's... I didnt add a power amp in, but i did mod it to sound much better.  I posted in the marshall workbench forum but thought i'd repost here since it comes up in google search.  One thing i forgot to put in that post - i clipped c10 too.  I didn't add a mid knob as it seemed good stock.

------------

This is basically a Princeton reverb without reverb or tremolo and it gets love as a pedal platform amp. It is much better than the TSA30. You can read about my struggles with learning about that somewhat odd circuit if you search for it. But the 15 is straight forward and is just kept super clean with that 47k cathode resistor on v1b. I have had this amp for years and it is a good pedal platform stock, but after all this time I found the clean tone to be sterile and bland. And pretty quiet. I ran the boost always on and it still barely got dirty at max volume. Isn't a TS's best sound in front of a breaking up amp? Not really possible here. So I decided to switch it up.

1. Power tube Grid Leak resistors - 470k is more than double normal fender so I swapped these.
r16 and r10 - 220k

2. Grid stoppers - with the gain of the preamp bumped up (see below) it will be possible to distort the power amp so I added grid stoppers while I was in there.
Lift the yellow wires running to pin 5 of both 6v6's. Put a 1.5k grid stopper resistor (or higher, 1.5k is standard fender) in series running to both pin 5s. I made a small daughter board from perf board for this to keep it somewhat sturdy.

3. Preamp cathode resistors - I will attach a bad hand drawing if you want to see what i did here and on the NFB. I wanted a comparison to stock. I recognize changing the power tube grid leak will mean it is not stock anymore anyways, but i can't resist adding some switches...
r19 - on - off - on dpdt for 1k (stock) - 1.5k - 1.5k +10uf bypass cap (Princeton)
r20 - on - off - on dpdt for 1.5k - 47k (stock) - 1.5k + 10uf bypass cap (Princeton)

4. NFB added on a switch - most complicated but gives you more typical blackface fender tone vs a pseudo tweed tone without it.
Lift ground side of r26 and c17. Connect them and run a 47ohm resistor from that connection in to one of those ground holes. Run a wire from the 8ohm speaker output. Connect 820ohm in series then connect that to the connection point of c17, r26 and the new 47ohm resistor that goes to ground.
I did this all on a dpdt on - on switch so I could get a more tweed-ish tone if I want. See my bad drawing to see how.


Last 2 that I haven't done since it sounds great as is now but could get it even closer to fender.

5. You could also swap the PI long tail values to be what they are in fenders, but i didn't mess with that yet and I might not since it sounds great now. Anyone want to describe what that rebias would do in the phase inverter? Keep it clean longer? Looking at deluxe schematic, R12 and R26 would be 470ohm and 22k.

6. Tone stack - the 470pf treble cap at C1 gives it a slight bump in the upper mids that traditional fender 250pf wouldn't have. I left it now because it sounds good that way, but I may swap it out since I have one on hand just to see the difference.

[bopelado]

1. Power tube Grid Leak resistors - 470k is more than double normal fender so I swapped these.
r16 and r10 - 220k

change for low Grid Leak Resistor, did you keep Bias at 30V? or did you change bias value?