Troubleshouting original Script Phase 90 .... Now Fixed!.

[Melanhead]

Well, I ve been playing with this thing for a while ... Basically had just the board and wired it up following RG's schematic:

http://www.geofex.com/FX_images/p180.gif

Well it worked for a bit once I set the bias properly ( although didn't sound quite right ... ) then I put the back cover on ... No sound! ( DAMN! SHORTED SOMETHING SOMEHOW! )... took it off, had sound but no longer phases ... found a bad IC's ( lm741 ) and a bad 2n5952 , replaced 'em ... still no phase

Here are some voltages ...

The Zener is at 4.88v

Looking at the schematic I called IC1 the first one top left ...

IC1

1 4.7 mv
2 4.87 v
3 4.64 v
4 0
5 N\C
6 9.07 v
7 4.86 v
8 4.9 mv

IC2 to IC5

1 3.8 mv
2 4.86 v
3 4.86 v
4 0
5 N/C
6 9.07 v
7 4.89 v
8 3.6 mv

IC6 Bottom Middle ( modulation section I believe )

1 6.1 mv
2 .4 mv
3 5.66 v
4 0
5
6 9.07 v
7 8.37 v
8 .3mv

all 2n5952

D 4.87v
S 4.88v
G 1.26v

2n4125

E 4.88 v
B 4.30 v
C 3.03 v

I'm aware that the 5952's have to be matched, could they have drifted ?? Also I did replace one .... Could this be part of it ?

Any help would be appreciated as I getting an understanding of how this thing works but it's very slow going

[petemoore]

Check out GEO Improved Jfet Matcher, or Jfet Matching Fixture...
 Yes, the Jfets should be closely clustered, all four of them, so...I don't know if the Vgsoff can be measured in circuit..you can either go by the 'other' ones that are in there...[pull one or two more and measure them to find where the factory Jfets 'fell' as far as Vgsoff...but ya never know unless you test them, how 'close' they are to one another.
 I bought 100 Phaser Jfets, matched them, got a six set of them matched, another set of four, also a set of two...you're not looking for a certain' number for VGsoff, but you want four Q's that are close to the same reading. I swapped around with the matched set of six and found four I liked for the 90...close enough gets good phase...
 So...consider socketting the transistors, that way you can mess with them, see if you can use the old ones...I'd just match new ones.

[niftydog]

Assuming all op amps are LM741s, and R.G.s statement at the bottom of the schematic is correct (and it looks that way to me) they should all have the battery voltage appearing at pin 7.

Only one of the op amps does.

[robbiemcm]

Noob question:

Okay... setting the bias propperly... I thought the schematic has this sussed out for people already?

[Melanhead]

Check out GEO Improved Jfet Matcher, or Jfet Matching Fixture...
 Yes, the Jfets should be closely clustered, all four of them, so...I don't know if the Vgsoff can be measured in circuit..you can either go by the 'other' ones that are in there...[pull one or two more and measure them to find where the factory Jfets 'fell' as far as Vgsoff...but ya never know unless you test them, how 'close' they are to one another.
 I bought 100 Phaser Jfets, matched them, got a six set of them matched, another set of four, also a set of two...you're not looking for a certain' number for VGsoff, but you want four Q's that are close to the same reading. I swapped around with the matched set of six and found four I liked for the 90...close enough gets good phase...
 So...consider socketting the transistors, that way you can mess with them, see if you can use the old ones...I'd just match new ones.

Thanks ... I thought it may be something like that ... I have another phase 90 board here for parts that someone replaced ( a very hack job! ) most of the 741's in ... I may try using all the fets from this board as they would have been matched at one time ... that's were I got the replacement from... If not ... I'll build R.G.s fet matcher and try that... Thanks muchly!

[Melanhead]

Assuming all op amps are LM741s, and R.G.s statement at the bottom of the schematic is correct (and it looks that way to me) they should all have the battery voltage appearing at pin 7.

Only one of the op amps does.

Isn't 7 the output  and 6 the V+ or am I missing something ?

[Melanhead]

Noob question:

Okay... setting the bias propperly... I thought the schematic has this sussed out for people already?

there's a trimpot to set it within a certain range, it's trickey as these FETs want everything just right or it won't work! ... I tweaked it when I first got the pedal wired up, basically turned the trimpot to where I heard it phasing... there's a small range that it'll work in and it sounds slightly different within this range ... and of course it seems that the sun and moon have to be aligned just right as well .... Not much of  a technical answer but I don't know the theory behind it ...

[niftydog]

Isn't 7 the output and 6 the V+ or am I missing something ?

the latter. check it out.

[Melanhead]

Thanks! ... I must have read the pinouts wrong and therefore wrote the wrong pins down ... I'm thinking I read it from 1 to 4 then jumped to the top right corner thinking 8 was 5 etc ... at least that's what I'm hoping ...

I'll have to check it out tonight....

At least it's another lead ... ahhh I love detective work.

[Melanhead]

Well finally got around to playing with it some more ... put in a newly aquired matched set of 5952's ( Thanks Scott! ) at first I put 'em in backwards ... It seems that the original pinouts of the 5952's don't match the Fairchilds! ... Got that solved ... Still no phase ... Replaced the pot 'cause as I was playing with it i could get it to cut in and out ... Presto!

FIXED! .... yahoo! ....

Thanks for all the help ... now I can move on to the next project ... Fix a script MXR Blue Box ....

I'm sure it'll be easy compared to this phaser ...