Teisco Checkmate 21 Solid State amp schem + help

[galaxiex]

Yep, as suspected current gradually goes up after 20 seconds of playing and then spikes to O/L on the meter.

As it does this there is a quiet hum that starts and builds but does not get very loud at all.
At that point I'm quickly switching the amp off.

Now to try and pinpoint the cause....

BTW the 2SB175 and 2SB324 transistors are available online but may be of questionable quality...
I may be able to sub 2N404 for the 2SB175?

[galaxiex]

Yes I think you are correct. 

I just went in and replaced ALL the 16V rated electrolytic caps for 25V or better rated caps.
The exceptions were the three .47uF caps that have 25V rating so I left those.

The amp is easily twice as loud as before,
I actually had to turn down the volume control and at "3" it's still a decent bedroom volume.

Still fades out after 20-30 seconds of playing.

A "real" amp inside the cabinet is still on the table as an option....
but I'd want it to have reverb and tremolo working too.

Got any schematics or donor amp candidates in mind? 

Or... maybe I can find some substitute output transistors...?  Re-bias likely needed.
There's gotta be some higher power rated PNP germs out there? Probably expensive...
Or silicon that would work?

I can see the next limitation being the phase transformer....  :blank:

[galaxiex]

HAH!
I found a pair of 2N2148 PNP Germanium T03 package in my junk. Way overkill I'm sure.... 

Quick and dirty wired them in with alligator leads (removed original 2SB324's of course)...
and it works!

Sound is a bit distorted and need the tone control at "0" else it squeals.
Maybe because of the long leads to the transistors?
Or biasing issue. Or both.

No heat sinks on the 2N2148's, they got barely warm after 5 minutes of playing and the amp did not fade out!
Yay!

Voltage stayed up around 24-25V. Didn't check the current.

I think I got some T03 sockets around here someplace.... 

[galaxiex]

Alright, I got the new transistors mounted.

The amp is sounding better than ever.
It's got some volume, tho not earth shaking.
Bass response has improved (which was another thing I hadn't mentioned yet, and wanted to fix).

The tremolo now has an interesting "pulse" or "throb" in behind the note that sounds kinda cool. 
I like it!

The tone circuit does need some attention because at 10 the sound has a harsh ice-pick distortion.
Tone at 4-5 and it sounds great!

Check out the pics and let me know what you think.

Now to get back to working on the reverb...

[galaxiex]

Roly, in your circuit above did you mean to draw the transistors Q2 and the added one as PNP type?

I have some 2N3906.  Suitable?

[Roly]

Doh!   :duh



1) for letting the blood rush to my head.  Sorry, just think, "add an emitter-follower".

2) getting the wrong datasheet and thinking "silicon".

That "heatsink" is a joke.

Grab a fan-blown CPU cooler and thermally couple it to that.

You could try increasing R25 which feeds the bias chain, say try 2k2 and see if that cools it down a bit, may also reduce output power a bit.

Try to avoid running it into thermal runaway - it isn't healthy.

The driver Q4 might need some thermal attention too.

Get some heatsinking in there!  As a germanium amp these really don't like heat.

{sorry again  } 

[galaxiex]

Doh!   :duh

1) for letting the blood rush to my head.  Sorry, just think, "add an emitter-follower".

Ummm ok, no apology needed. I can figure out an emitter follower.

2) getting the wrong datasheet and thinking "silicon".

Again, ummm,
in reference to what? The emitter follower?

That "heatsink" is a joke.

The stock original heatsink?... or the brackets I made for the 2N2148's ?
Because the 2N2148's run quite cool and don't even need heat sinking.
The brackets merely hold them.

Grab a fan-blown CPU cooler and thermally couple it to that.

You could try increasing R25 which feeds the bias chain, say try 2k2 and see if that cools it down a bit, may also reduce output power a bit.

Try to avoid running it into thermal runaway - it isn't healthy.

The driver Q4 might need some thermal attention too.

Get some heatsinking in there!  As a germanium amp these really don't like heat.

{sorry again  }

Again, no apology needed but I am confused....
Are you suggesting I put the original 2SB324's back in with proper heatsinking?

For R25, I don't really want to do anything to reduce the output...
...unless you think it's really needed with the 2N2148's.

Thanks for the data sheet, that's the one I found too. 

Are you ok Roly?
The above was not your usual clear concise post.

[J M Fahey]

Maybe he already got sick of this endless thread.
Sorry.

[galaxiex]

Maybe he already got sick of this endless thread.
Sorry.

I would not be surprised...

Am I annoying everyone with this?

[Roly]

Not at all, I'm finding it "interesting" to try and go back lebenty-leben years when I was wrassling this sort of gear on the bench, where the red wire was actually -22V; and this "asian" PNP format is still really confusing to somebody brought up in the British school of circuit drafting, voltage rails stacked by polarity/potential and the stages strung in between.  {Worked on Decatron pill counters and such that had several supply and bias rails stacked up from most negative on the bottom.}

The previous was intended to only be diagrammatic with that in mind, you need to be mindful of the details (and well-intended bum advice).  So my excuse is that the drive amp representation is "conceptual" - okay?   

[galaxiex]

Okay. 

With that cleared up...

I've been playing this amp for awhile now with the 2N2148's and it seems to be good.
With no effects in use the amp stays clean to max volume as long as the tone control is not above 6-7.
I have some ideas for that...

I've cobbled up a separate 5V supply for the reverb with the ground of that going to ground (+Ve) on the amp.
No more hum, but I did not yet put in an emitter follower.
I have C11 dialed completely out on Q2 and it's playable at low reverb settings.
Much past 4 and reverb gets over the top and all goes to mush.
At 10 its on the edge of oscillation/feedback and does oscillate if tone gets turned up.

All this stuff is interacting somewhat of course.
I expected that.

So 2 things... need to tame the tone control and get the reverb levels sorted.

I'll do the emitter follower first and see what happens,
but for now gonna take a break from wrassling with this thing.

I AM gonna wrassle this thing to the ground and make it beg for mercy.

Some might call it stubborn...

I call it dogged determination.  :duh 

BTW just a point of interest...
The reverb module specs say 100mA... but actual testing show it drawing @ 167mA.
With that in mind and to have something to compare, I have ordered the "newer" small reverb brick BTDR-2H labeled Accutronics rather than Belton.
Probably both made in the same factory somewhere in Asia anyway....

[Roly]

Got get'um Tiger!   :dbtu:

actual testing show it drawing @ 167mA

Woah,    that is seriously unhelpful; that's a almost three times "typical", and 50% over "max".

{I hate that, when the datasheet lies.  :trouble }

[galaxiex]

{I hate that, when the datasheet lies.  :trouble }

Ya, not sure if the data sheet is lying, or something is wrong with the one I got. 

That's why I ordered another one. 

[Roly]

hmmmm 

These digital sorts of things tend to work or not, no half measures, (however ...)

[galaxiex]

Ok, Had a nice break... back to it. 

And back to basics...

I came across this thing called SWTC, (Stupidly Wonderful Tone Control), and some modified versions of it.
In the schem I have used the SWTC 2.

Here I have stripped down the schem to "back to basics" to get the basic amp sounding good.
I didn't like the tone control and how it caused the amp to sound bad with tone on 10.
Also the volume control was not very "linear" so to that end we use an audio taper pot. 
Besides, the original pots were still a bit scratchy even after using spray lube/cleaner.

I've not yet made these changes to the amp but will do so tonight after work.

After I get the basic amp sounding good, I will add back in the reverb circuitry.

Note that compared to the last previously posted schem, I have eliminated Q2 and its associated parts.
Q2 was the original reverb driver which we don't have to use,
and now have the option of using it for an added gain stage or whatever.
The reverb can also be tapped into the signal chain at whatever point we want, but that's all for later...

I have also been considering trying different (silicon) transistors for Q1-3-4 but again... later...

This is becoming an almost complete re-engineering of a cheap amp that would have been easier to just gut the thing and put a "real" amp in it.

I don't care, I'm having fun and learning lots.