Troubleshooting some odd LND150 circuit behaviour...

[mdcmdcmdc]

Hi folks,

I'm wondering if anyone might be able to help me figure out an issue I'm having with this circuit I put together.

Here's the schematic:



The circuit worked well on the breadboard, but i'm experiencing some odd behaviour when it's built up on a PCB.
Essentially, it initially worked as expected, everything seemed to be great, and then all of the sudden it sounded starved/thin. The voltages on all of the fets and bjts remained constant but it was as if the circuit was somehow being throttled at the input. So, very little distortion, kind of farty/crackly sounding.

This had happened once or twice on the breadboard, but I assumed it was just a bad connection and jiggling some parts around or disconnecting and reconnecting things cleared it up.

Similarly, I had initially socketed all of the LND150s and BC549s on the PCB, so I swapped them all out, no change, then swapped them again and it went back to normal. Everything worked great and I assumed it was a loose connection in the socket maybe, so I soldered in the transistors. And, lo and behold, the problem has come back.

I've double/triple checked all of the voltages, I've reflowed the whole board, everything seems a-ok, and when it was working it was solid - ie, jiggling things around had no effect on the circuit, it didn't seem like a loose part or anything.

The normal troubleshooting technique of just kind of poking the back of the board hasn't yielded any results. I'm assuming this is a design flaw in the circuit or there's something about the LND150 that I don't understand... but I'd love to figure out what's going on.

Thanks!

[mdcmdcmdc]

Hmmmmm... might be a wonky trimmer on Q1 drain. I followed the board with an audio probe and things were going sideways between the D and G on Q1. Clean signal into the gate, weird distorted mess coming out. Gave the trimmer a few turns back and forth and it seems to be ok again?? Doesn't make a lot of sense as the voltages were fine though...

[mdcmdcmdc]

HMMMM.. the plot thickens!
Built another PCB up, soldered the transistors/FETs directly to the board this time, fired it up... same issue.
Poked around with the audio probe, wiggled the trimmers a bit, voltage on Q2 drain jumped a few volts, now everything seems to be working as-it-should.

I'm curious if there's anything in the first stage that could be holding voltage? is there a cap that should not be? Again, I put this together by ear/trial and error so I'm sure there's a simple enough design flaw that someone might be able to spot. It's very odd.

[iainpunk]

your problems seems like static on Q2 gate, it needs a pull down

1M should be OK

cheers, Iain

[mdcmdcmdc]

Found a nice place to tuck a 1M on the bottom of the board and the problem has not resurfaced - hopefully that solves it. Thanks so much! I love an easy fix.

[iainpunk]

i strongly believe that is it.
the Gate pin is like a capacitor, no DC can run through that pin, and the components connected to that pin also don't conduct DC, so if a voltage somehow makes it there (and oddly enough, that often happens) the floating DC voltage on a mosfet pin is like somethings changing a BJT base current, but yours is uncontrolled. this creates a misbiassing of the gain stage, you described its sound as how i would describe a misbiassed transistor.
if it occurs again, maybe lower the resistance to 470k for a faster discharge.

cheers

[duck_arse]

the lnd150 confused me before, is probably still confusing me. it's

N-channel depletion mode (normally-on)

-edness in particular. is it possible the DC path to ground on Q1 is buggering it up in the same manner no DC path on a jfat gate buggers it up?

[iainpunk]

the lnd150 confused me before, is probably still confusing me. it's

N-channel depletion mode (normally-on)

-edness in particular. is it possible the DC path to ground on Q1 is buggering it up in the same manner no DC path on a jfat gate buggers it up?

a LND150 and other similar mosfets are to be treated like Jfet's, when it comes to such guitar circuits, a bit less gain, and they can die of static, but the rest is the same-ish

cheers