Jfet matching for phase 45 build

[Locrian99]

Finally got around to trying out 1n5231 diodes the 500mw 5.1v.  Results are great, going with this one.   Took a bit to get there but pretty happy with this now. 

[Mark Hammer]

I should have mentioned it before, but there is nothing "magic" about the 5.1V zener.  The thing is that the bias voltage applied to the FETs is divided down from the supply voltage.  When players mostly powered their pedals with batteries, the challenge was to provide a *fixed* bias voltage, even as the battery voltage began to drop.  The zener assured that as the battery dropped, the bias would be derived by dividing down a stable 5.1v.

My guess is that, in an era when most pedals are externally powered, one could comfortably omit the zener, although the trimpot setting would be different because you're dividing down 9v.

There are probably more things we perceive as critical or necessary, which were based on the presumption of battery power.

[Locrian99]

I should have mentioned it before, but there is nothing "magic" about the 5.1V zener.  The thing is that the bias voltage applied to the FETs is divided down from the supply voltage.  When players mostly powered their pedals with batteries, the challenge was to provide a *fixed* bias voltage, even as the battery voltage began to drop.  The zener assured that as the battery dropped, the bias would be derived by dividing down a stable 5.1v.

My guess is that, in an era when most pedals are externally powered, one could comfortably omit the zener, although the trimpot setting would be different because you're dividing down 9v.

There are probably more things we perceive as critical or necessary, which were based on the presumption of battery power.

Yea i had read your thoughts on that on a previous post.   Actually quoted it earlier in this thread around when the 78l05 got brought up as well.   It may just have been in the biasing but I noticed at or maybe just perceived some clipping as it drifted further from the 4.5v when it was biased with the trim setting at where I felt it was doing the max phasing. 

[Rob Strand]

I should have mentioned it before, but there is nothing "magic" about the 5.1V zener.

The 5.1V zeners biases the opamps for best swing.  A higher voltage will increase the swing of the BJT stage - which hasn't got a great swing to start with.

The choice of zener voltage should be for clipping as the trimmer can take up the slack for the JFETs.

On average the 5.1V zeners have the lowest low drift of all the zeners.

Actually quoted it earlier in this thread around when the 78l05 got brought up as well.   It may just have been in the biasing but I noticed at or maybe just perceived some clipping as it drifted further from the 4.5v when it was biased with the trim setting at where I felt it was doing the max phasing.

That sounds a bit odd unless the clipping wasn't clipping it might be some of the JFETs are getting cut-off.
Measure the 5V rail when you first power up the pedal, then play the pedal to see if it is clean.  What about 10mins
then see if you can hear any "clipping" and remeasure the 5V rail.

The 78L05's can drift as they warm up.   I'm a little surprised you could detect it on a phaser though.  On some measuring equipment I've replaced 78L05's with TO-220 packages LM7805 to reduce power-up drift - even though the circuit was only pulling 10mA or so.

For general drift the 78L05 should beat pulling a random 5.1V zener out of the box and using it for a phaser but the zener might win on *warm-up* drift.

Maybe look elsewhere or take some measurements of the 5V rail before going down the drift rabbit hole.

[Locrian99]

I am not using the 78l05, I was referring to earlier in the thread it was mentioned instead of the zener.     With the different zeners I was referring to when they weren't supplying in the ballpark of the 4.7 I noticed some clipping, well at least what I perceived to be clipping who knows, it sounded best to me with the 500mw 5.1 compared to the other 4 ways I tried it.