GCI Apostle - Troubleshooting

[idy]

Red probe first? so G-D is Red on Gate? no, you are probably doing Black first, so
G-D= black on G...

But yes, that looks better now.

What is the pinout of your PCB?

I will look and try to figure it out...

[idy]

That way, they look like JFETs. Can you describe your JFETs to me now, facing the flat part, legs pointing down they are left to right...D, S, G?

Now ask, what is the pinout on the PCB? Of the three pads, which one goes towards ground and which towards power? And the gate, which pad is attached to R3?

Do these two things match?

[pangeadestructor]

That way, they look like JFETs. Can you describe your JFETs to me now, facing the flat part, legs pointing down they are left to right...D, S, G?

I see what you're getting at now, yes facing the flat part with legs down it is DSG. The pinout is definitely correct on the pcb, bottom end of r3 is connected to the gate on this jfet.

[idy]

And you still see those same voltages on the legs of Q1 on the board, around 16v on all 3? With the number falling slowly when you touch them with the meter?

[duck_arse]

what's all this resistance business? the datasheet will tell you what you need to know pins wise.

and then, your voltmeter, measuring the fet pins in-circuit voltages, will tell us more about what is going wrong/on.

[pangeadestructor]

and then, your voltmeter, measuring the fet pins in-circuit voltages, will tell us more about what is going wrong/on.

Yes, I currently have that trimmer dialed all the way down but am getting the same voltage at all 3 pins of the q1 jfet when I power the circuit - 1.698v. If I turned that trimmer all the way up, it would likely read somewhere around 15.6v.

Although when I first plug it in after leaving it unplugged overnight, the jfets will all be able to go up to ~36v using the trimmer test points for each one. I haven't been able to plug in, get that full expected range of voltage up to 36v, and also measure each of the legs during that period before my jfet max voltages all begin dropping.

[pangeadestructor]

bumping this, thanks

[idy]

I always warn against random pulling of components but...
At this point I start to think it might be OK to remove Q1 and put a socket in there.
We know those really are JFETs, and the pinout is correct. But you imply that the 2n5457 you tested was from a different source? So no information there...
But if you are seeing the same voltage on all three legs, that is not good.

2n5457 is not designed to take 36 volts. Data sheet says Max voltage D to S: 25v. You might have killed them all. Yes?

So it might be worthwhile to remove Q1. Maybe even '"sacrifice" it by cutting it out, the point being avoid damaging the pcb.
Then put a socket in there. And measure, with no JFET, the voltages. But if they are all fried, maybe all need to go. I really hope someone else will pipe in. I hate to advocate for pulling components from a PCB.

I am hoping someone else has a better idea. I suggested breadboarding that first stage...

A voltage that falls over a period of time suggests a capacitor...

[duck_arse]

And measure .... the voltages. ..... I really hope someone else will pipe in. ...

and when we say measure the voltages, or what are the voltage readings, what we really really want to see is all the voltage readings on all the pins of all the jfets, written, shown, posted, in one place. set yer bias however you like, but then do the measures and report those readings. please.

[pangeadestructor]

I'll measure the voltages across all jfets as soon as i have time, before ripping anything out. Thanks, I really appreciate the assist here.

I did have some concerns that I had fried the jfets by leaving the trimmers turned up on the initial power up, since those allows them to go up to 36-38v, hoping that's not the case.

[pangeadestructor]

Must have fried at least one of the jfets, I pulled out q1 and replaced it with a new 5457 and voila there's output.

It's still very quiet and the Boost control appears to work backwards. The audio is very loud going into r13 and very quiet coming out but the values are correct in that area and the solder joints look good. I'm thinking q3 needs a replacement as well, voltages on that are D 0.33, G 17.72, S 17.73.

[idy]

Congratulations.

So you hear boost from first, replaced Q1.
And more after Q2,
But things die again at the gate of Q3, just like used to happen to Q1?

And you are sure about those voltages? People here die of sadness when they don't see voltages for all Qs posted every time things change. It is a custom here among the natives, we show each other our voltages as greetings. Are you sure that: the D (which usually goes towards the power supply) is lower than the S, (which is usually attached to ground?)

I would think time to replace Q3. Examine and make sure there is not a short on PCB between G and S... but dollars to donuts...

Sorry no one noticed that data sheet says 2n5457 only takes 25v. Hey guys, is this common, putting 25v FETs in a circuit with 36V ....and trimmers? Going rogue? Risk taker... It's a jungle out there.

[PRR]

> ...putting 25v FETs in a circuit with 36V ....and trimmers? Going rogue? Risk taker... It's a jungle out there.

A JFET on a 36V supply normally won't sit with full 36V on it. Gotta leave some for the resistors.

UNLESS those resistors are trimmers and someone has been hitting the endstops. Obious if the plate resistor is zero that leg of the FET will be full 36V. Since the other channel leg is at 1V-2V, you can have >30V.

But also note that Vds on a JFET may not be a blow-up number but a fails-spec number. Gate leakage rises past 10V d-s and by 30V d-s BJT are competitive (and cheaper). However 36V is the nominal limit for many analog processes (see all the old jellybean opamps) so best not flirt with 36V.

[pangeadestructor]

I totally understand the frustration when someone asks for help here and doesn't provide the basic info like all voltages. Just wanted to update in case anyone experiences a similar issue and this might be helpful.

It was definitely the jfets, q1 and q3 got cooked with the initial 36v when I first built and tested the circuit. I was surprised to find the first time that the jfets in this circuit will indeed go up to the full 36v with the trimmers all the way up, which is well past the limits on their data sheet.

I replaced q3, dialed all of them to around 17v, and this thing is an absolute monster.

[eh la bas ma]

Glad you managed to sort it out. Congrats !

On my build, the Mode toggleswitch tends to be dodgy and microphonic.

I checked with a DMM, and either R6 or R11 are grounded depending on the toggle position, as it should be, if i understand the schematics correctly.

On the right position there is a slight mid boost compared to the left position, and that's all i can notice.

I wonder what does it do on your build ? Just a subtle mid emphasis too ?