News:

Forum may be experiencing issues.

Main Menu

Phase 90 - I'm a n00b

Started by daleykd, July 29, 2017, 03:42:45 PM

Previous topic - Next topic

daleykd

I was blessed by someone and was given a beautiful 1590A Phase 90 PCB.  The Q2-Q5 FETs were SMD MMBF5458.  This is my first Phase 90.

I got it all soldered up and started trying to bias by ear, and there's only about a 0.0000000001 spot where I get much of a sweep.   I DO get a sweep, but it's not very intense like I'd expect.

I was told that I don't have to match the SMD JFETs because the tolerances are much tighter.

I also have a Rigol Digital O'Scope, but I have no idea how to use it to bias.  (I would LOVE it if someone had a tutorial on how to do this.)

Am I completely missing something here?  Maybe I really DO have to match the JFETs?  Maybe I just have to find a sweeter spot that's more like 0.000000000000000000001?

(I've had too many boards not work lately.  I hate being in a rut.)

jimilee

Sounds about right, it's a tight space that it sounds right. It's my understanding smd are fairly close as well. Is it the script version or the block version?


Sent from my iPhone using Tapatalk
Pedal building is like the opposite of sex.  All the fun stuff happens before you get in the box.

daleykd

Quote from: jimilee on July 29, 2017, 03:50:45 PM
Sounds about right, it's a tight space that it sounds right. It's my understanding smd are fairly close as well. Is it the script version or the block version?
Tight space is an understatement.  :)  There's a switch called SCRIPT that does script vs. block (I think).  Flipping it makes it so that it squeals at the "peaks" of the sweep.

jimilee

Something isn't right for sure, but I wouldn't think the fets would do that.


Sent from my iPhone using Tapatalk
Pedal building is like the opposite of sex.  All the fun stuff happens before you get in the box.

daleykd

I tried to reflow the FETs.  The first time I did, it totally changed where they turned on with the TRIMMER.  Then, I reflowed them again, and the position was back to about where it was, and the turn of the trimmer felt better.  Instead of microscopic turns, just minimal turns was affecting it.  However, it still just isn't very phaser-ish.

I'll try another set of MMBF5458 tomorrow or whenever.  I'm too frustrated because I don't even know where to begin.

bsoncini

#5
Have you read this to understand the circuit better? Should help you to know where to probe with your scope and audio probe.

http://www.electrosmash.com/mxr-phase90

somnif

This fellow has a method of visualizing a phaser with an o-scope: http://www.davidmorrin.com/home/trouble/troubleeffects/phasers  but it requires a swept function generator. Having never played with a scope I have no idea if this is helpful at all, but there you go.

Also, annoyingly few people have reported on matching SMD jfets. Frustrating, that.

alanp

Call me Mr Cynical, but my first suspicion would be badly matched FETs.

Transistors, you can get away with a few from the same tape. There is the odd outlier, but on the whole, transistors tend to be close.

FETs are flagrantly aberrant.
"A man is not dead while his name is still spoken."
- Terry Pratchett
My OSHpark shared projects
My website

wgc

Based on your post about reflow, I think you may have done a little heat damage to your jfets, which made them less of a match. Possibly you have a wrong part value somewhere.

This guy posted some results about tolerances of smd jfets vs thru hole (near the bottom).

http://surfybear.weebly.com/blog/the-jfet-tester

When I did my phase90 smd layout, I used two dual jfet packages for the jfets. I did no matching, and both came off the same tape. I built two boards, both worked great with very little adjustment to the trimmer.

I would expect the tolerances within the same packaged pair to be very close unless one side got a little fried. I would expect more variation between packages but still probably close enough if they're off the same tape.
always the beautiful answer who asks a more beautiful question.
e.e. cummings