VGs off - LFO swing

[armdnrdy]

I recently dove back into a project that I was working on that was never quite right. Jet Phaser. This is my second stab at it.

I have a problem at the bottom of the sweep. This is the same problem I had with the first build. It seems as if something shuts off, as if the phasing falls out before turning around and going back up. It sounds for lack of a better word...sloppy! If I adjust the bias up, out of the problem area, it doesn't sound like the sound samples on youtube. It sounds thin and airy, instead of thick and gutteral.

I built another board that is a trace from the original, I've sourced all original components, I've used the service manual for reference, as well as pictures of both sides of the board to verify the schematic and 99% of the components. What gives?

I started taking voltages when I came upon a question that somebody might help me with.

Where do the Jfet VGs off have to be in relation to the minimum LFO swing voltage? I'm thinking as close as possible.

Any help would be appreciated

[R.G.]

No problem. You just need germanium JFETs.

On a more serious note, this is what I was running down when I did the work on the Phase 90 JFETs and the JFET matcher.

Your LFO must be able to sweep the JFET's channel resistance through a useful range for the circuit - whatever that circuit demands. This differs from circuit to circuit.

A JFET's Vgs changes the channel resistance from rdsOn to infinity (or nearly so) over the range of Vgs=0 (for rds = rdsOn) to Vgs=VgsOff (for rds = infinity). So your LFO voltage range should be less than VgsOff from peak to peak. It *may* need to be a lot less, depending on what resistances the circuit needs. For instance, if your phaser only sweeps an audibly meaningful frequency range for phase resistors from 2K to 100K, then you only need to sweep the JFETs over that much.

... and now you have offset problems. If the JFET sweeps mostly in the non-useful ranges for the sweep of your LFO, you get audible dead spots. So you not only need to get the LFO center in the range that makes the JFET rds make the circuit sound good, you also have to scale the LFO peaks to keep it in the "good" range.

What gives is that JFETs vary - a LOT. This classic problem with JFETs has persistently been their downfall. We try to cope with measurement and selection.

[armdnrdy]

Thanks for the explaination R.G.

I think this phaser goes a bit beyond the normal matched jfets. It seems to be a bit particular about range of VGs off as well. Some time ago I built a modified phase 90 circuit (extra switchable phase sections) and had no problem dialing it in.

This thing.....I'm hoping to get it the second time (build) through!

[R.G.]

I think this phaser goes a bit beyond the normal matched jfets. It seems to be a bit particular about range of VGs off as well.

I think that's true. I've never messed with this specific one, but I seem to remember others having problems with it. It could well be finicky about several things.

[boogietone]

No problem. You just need germanium JFETs.

This made my rather bad day better. Thanks!
;-)