Modified Heathkit TA-28: Help me understand this

[anotherjim]

I don't know if I've nailed it. I was thinking more of seeing how it behaved with a buffer in front.
There should be 100nF DC blocking cap on the circuit input so I don't totally believe it could screw the bias.

I'm more worried by the 50k input level pot load and wonder if Frukost has wired it for true bypass or the original tone-sucker bypass.

[pinkjimiphoton]

even with the blocking cap, the ac <guitar signal> will go thru. sometimes the input signal is enough to make it change. what do i know lol

[Frukost]

I haven't checked yet, but it struck me today that I may accidentally have switched the transistors place while soldering. Would it be feasible that it could have caused the symptoms described?

(It's hard to see since the transistors printed surfaces are soldered face to face. Here's the veroboard again: http://3.bp.blogspot.com/-a_OghF-Jli4/UGSLSHyaBuI/AAAAAAAACuc/fhOMNCFPEZQ/s1600/Heathkit+TA28+with+trimmer.png).

Happy new years everybody!

[pinkjimiphoton]

yeah! that would do it! lol

swap em around, and let us know!   

[iainpunk]

I haven't checked yet, but it struck me today that I may accidentally have switched the transistors place while soldering. Would it be feasible that it could have caused the symptoms described?

no, not really, if you switch the transistors out, youll get a very sh1tty and lo-fi buffer-like unit probably way less volume than clean would have.

cheers

[pinkjimiphoton]

well,. MAYBE.
SOMETIMES at certain voltages and biasing, a reverse beta q will still work, and sound great in a fuzz. more common in ge than silicon, but it happens.

jackie orman has an article on using reverse beta'd q's on AMZ, you can most definitely get some weird stuff to happen that sounds really musical with stuff wrong sometimes, at least in distortion circuits.

i've encountered the tone knobs working backwards before over the years, i forget exactly which circuits exhibit this, but usually it comes from the bias being way off...

hard to get more way off than reversing a pnp and npn's orientation. and at a volt and a half, shit gets even weirder! it may stil pass signal.

i can't wait til he figures this out so we can have the truth revealed

[iainpunk]

well,. MAYBE.
SOMETIMES at certain voltages and biasing, a reverse beta q will still work, and sound great in a fuzz. more common in ge than silicon, but it happens.

jackie orman has an article on using reverse beta'd q's on AMZ, you can most definitely get some weird stuff to happen that sounds really musical with stuff wrong sometimes, at least in distortion circuits.

i've encountered the tone knobs working backwards before over the years, i forget exactly which circuits exhibit this, but usually it comes from the bias being way off...

hard to get more way off than reversing a pnp and npn's orientation. and at a volt and a half, shit gets even weirder! it may stil pass signal.

i can't wait til he figures this out so we can have the truth revealed

i think he means he put the npn in the pnp's place and vice verse, not that he turned them 180

those are two quite different outcomes.

cheers

[pinkjimiphoton]

true, but functionally it can be the same outcome. thats why you can run a neg ground pnp fuzz. depends on the circuit itself, but weird shit happens in very unexpected ways when dealing with the unlinearities involved in distortion.
i have definitely reversed q's before and still had it pass signal.
reverse beta i think is similar in the ultimate outcome.
probably wrong, i often am, lol... but i've pulled off some wack shit over the years

[Rob Strand]

i've encountered the tone knobs working backwards before over the years, i forget exactly which circuits exhibit this, but usually it comes from the bias being way off...

With 1.5V and 3V circuits it's very easy for the bias to be off.   Some circuits even changing a resistor one value in the right direction pushes the biasing too far in the wrong direction the other way.   The Heathkit at least has some DC feedback which helps reduce that problem.

It's probably best to play with R5 and R7.   

The game would be set R7 then tweak R5 to sound good.  Try another R7 then repeat.
Finally choose which pair of R5 and R7 is best.

Here's some examples where things could end-up:

R7      R5
8.2k   22k   
6.8k   33k   
5.6k   56k

[pinkjimiphoton]

thanks rob

@#$% it. i SHOULD be working on stuff for mla...

but i think i'm going in and revisiting this @#$%er.
if i know me, i'll set it up to run 9v

as i recall, a bunch of diodes in series to drop it down to 1.5 is possible. man, been years!
i never had much time with the one years ago, my co-guitarist lusted for it hard so i gave it to him.

stay tuned

[Frukost]

thanks rob

@#$% it. i SHOULD be working on stuff for mla...

but i think i'm going in and revisiting this @#$%er.
if i know me, i'll set it up to run 9v

as i recall, a bunch of diodes in series to drop it down to 1.5 is possible. man, been years!
i never had much time with the one years ago, my co-guitarist lusted for it hard so i gave it to him.

stay tuned

Did you check the circuit i posted? It runs on 9V, unless of course you'd rather step it down your way.

[Rob Strand]

@#$% it. i SHOULD be working on stuff for mla...

but i think i'm going in and revisiting this @#$%er.
if i know me, i'll set it up to run 9v

as i recall, a bunch of diodes in series to drop it down to 1.5 is possible. man, been years!
i never had much time with the one years ago, my co-guitarist lusted for it hard so i gave it to him.

I can't say the specific values in the table will make it sound like the "original" they can only be consider as something that biases OK.   The idea of stuffing around with R5 and R7 generally works.

[Rob Strand]

Did you check the circuit i posted? It runs on 9V, unless of course you'd rather step it down your way.

From what I can see the trim pot drops the voltage to the circuit to 1.5V.

For low voltages I actually prefer to use a shunt regulator with 2 or 3 diodes in series to ground.   Like version (c) here but with "cc1"  part as resistor (maybe 4.7k).  You can also use a single RED LED.

[pinkjimiphoton]

thanks rob

@#$% it. i SHOULD be working on stuff for mla...

but i think i'm going in and revisiting this @#$%er.
if i know me, i'll set it up to run 9v

as i recall, a bunch of diodes in series to drop it down to 1.5 is possible. man, been years!
i never had much time with the one years ago, my co-guitarist lusted for it hard so i gave it to him.

stay tuned

Did you check the circuit i posted? It runs on 9V, unless of course you'd rather step it down your way.

that way will work, but sometimes the current present could end up smoking the trimmer. dc on pots often leads to issues.

if ya put a bunch of diodes in series, you can use the voltage drops to get you the voltage you need. i think it was mark or rg that told me about that years ago.

so 9v to 1.5 would be, with a diode dropping .6 v,  so you'd need like, 12 914s in series or something to drop it down to 1.8v from 9.

easier to do a voltage divider to get 4.5v, then drop it down i guess...then 5 diodes would give ya almost exactly 1.5v.

or maybe my dope addled brain is having a brain fart

[Rob Strand]

so 9v to 1.5 would be, with a diode dropping .6 v,  so you'd need like, 12 914s in series or something to drop it down to 1.8v from 9.

easier to do a voltage divider to get 4.5v, then drop it down i guess...then 5 diodes would give ya almost exactly 1.5v.

or maybe my dope addled brain is having a brain fart

The problem with dropping the voltage so much is suppose you tune the diodes to get 1.5V output with 9V input.   If someone plugs  an adaptor with 9.5V the voltage now rises +0.5V to 2V.  And for a battery if the battery drops to 8V the output drops 1V down to 0.5V.  The output voltage is all over the place.
[Mark Hammer actually reported this as a problem in a post in the Lounge a couple of years ago.]

Dropping resistors aren't great but they can work better than the series diodes.

The shunt diodes on the other hand produce quite a constant output voltage over a very wide range of input voltages.

[pinkjimiphoton]

not sure i understand how the shunt diodes work.
i get the diode drop adding up thing, but you only get the drop from the one diode shunting it across the rails, don't you?

[Rob Strand]

not sure i understand how the shunt diodes work.
i get the diode drop adding up thing, but you only get the drop from the one diode shunting it across the rails, don't you?

It more or less creates a fixed voltage like a zener regulator.    (Like the zener on MXR Phasers.)
It's called a shunt regulator.

When you measure the voltage drop across a silicon diode you always get around 0.6V to 0.7V.  So the silicon diode is acts like a zener diode with a low voltage.   The same goes for an LED.   The difference between the zener and diodes is for the zener the cathode goes to the output (ie. it points up) but for the diode(s) it points down.

http://www.learningaboutelectronics.com/Articles/What-is-the-voltage-drop-across-a-silicon-diode


The final voltage depends on the number of diodes.

The only thing you need to do is to choose the series resistor low enough that the diode(s) always has some current.  So enough current has to go through the resistor to power the circuit and a bit left over to keep some current through the diode.

It wouldn't be suitable for a 1A power supply but for pedals with 10mA or less it works great.

The other way to think about it is the diode acts like a clipper.   It doesn't matter what the input is (the DC power supply) the output is "clipped" at one diode drop (or however many diodes you use).

With diodes as droppers the drop is a fixed voltage but the output is varies when the input varies.  With this shunt regulator the output is fixed and the drop across the resistor varies.

[Rob Strand]

John Hollis uses a similar trick on the Easy Vibe,

http://www.hollis.co.uk/john/easyvibe.jpg

See the top left corner.

[pinkjimiphoton]

isn't that kinda what i was suggesting? maybe i worded it wrong.
to be a shunt, its between the rails, right?
i think what i suggested was inline with the + rail.

i'll have to mess with it. thanks rob!

[Rob Strand]

isn't that kinda what i was suggesting? maybe i worded it wrong.
to be a shunt, its between the rails, right?
i think what i suggested was inline with the + rail.

Yes, that's the difference.  One is in-line (constant drop) and the other is a shunt between the rails (constant output).