What Does the Foxx Phase?

[anotherjim]

Trouble is there are limits to pot values. If you take 6.8M as a possible maximum, the variation with a single pot + a fixed resistor is limited. The pot value is too small compared to the fixed resistor to achieve enough variation of the total resistance.
Also, as I said, changing R34 also changes the DC level controlling the FET's, so you have to correct with the bias pot to re-centre the sweep. Not really practical. If R34 included a trim pot, you still have limited range because they don't come higher than 1M. IMO, the practical choice, would be to make R34 "SOT". This means "select on test". The resistor is replaced by header pins and you try different resistors by wrapping them on the pins. When you find the best value, you clip off the excess wire and solder on to the pins. The downside is that you'd need a good range of values between 2.2M and 10M to hand, which most of us don't normally stock. I do have them, so will try it that way.

[nickbungus]

After the post I started looking for high value pots and like you said, the 1M was the highest I could find.

Sorry, I forgot about the biasing.  I am new to this and a bit dumb/forgetful!

The SOT approach does make sense and seems to be inline with what has been previously discussed about different units having different values for R34.

[armdnrdy]

What I don't get is how the 1482 schematic is on the net and the studio isn't.  

Many schematics are not on the internet. I would imagine that the 1482 is....and the Studio Model 7 isn't because the 1482 was available for immediate purchase from Musicparts.com. As you've seen....you put in a request for the other Foxx phaser schematic over a week ago....still not available.

Especially coming from Greg Covington, as it seems he definitely did his homework.  His site is down now but he make all the Brian May stuff.

I've looked at an image of Greg Covington's May Z phase board. It appears as if he followed the schematic (Sears) and added a few obvious changes from images he gathered from the net. (C16, C21 .1µf, power supply filtering 12Ω resistor & 220µf capacitor) Greg's build includes the R25/C12 that is on the schematic but deleted from latter production units. (more on that next post)

[nickbungus]

Larry, your powers of research are incredible. Hats off!  It looks like you and Greg followed similar paths then.  You pro's sure know what you are doing!!

No, I never heard back from MusicParts.com, I sent him another email chasing him and no reply.

Really looking forward to your findings and schematic, and then building it!!

[armdnrdy]

Here's the schematic that I've drawn using the 1482 drawing and images of actual boards as reference.



PDF for higher resolution:

https://dl.dropboxusercontent.com/u/53299166/DIYstompboxes/Foxx%20Foot%20Phaser%20Mod.%207.pdf

Differences that I've found:

R1 input resistor is 100K instead of 270K
C16 & C21 are .1µf instead of .01µf
R13 is 300K instead of 330K
Rotary switch with two additional LFO speeds was added.
R25 and C12 were installed on Sears Phaser and early production Foxx Foot Phaser. These components were deleted in later production Foxx Foot Phasers and the Studio Model 7.
R34 6.8M instead of 2.2M
C18 changed to 10µf polarized capacitor instead of .01µf film
Power filtering added R53 (12Ω) and C25 (220µf)

I have verified R34 as 6.8M on many of the closer images. This is the best shot I've come across. (carbon comp resistor in front of IC3)


Here's an image of the empty holes where R25 and C12 would have been placed originally.


I've went through the images and put together a component overlay for the board values.
As you can see in the images, some of the trace routing can be seen from the component side due to the solder rosin soaking through the fiberglass board.
With this info...I have been working on a designation overlay to match the schematic. I'll post it when completed.

[Fender3D]

Sorry guys,
but...
won't TR1 and P1 feed DC voltage on output jack?

[armdnrdy]

Sorry guys,
but...
won't TR1 and P1 feed DC voltage on output jack?

I thought the same thing Federico.

Look at the original schematic on page 1 of this thread.

The only difference I see from that schematic and the actual board is: R51 comes directly off of pin 8 (output) of IC2C, then the output travels to the bottom of the board where it's connected to the positive side of C22. From the negative side of C22, the output signal is routed to the opposite side of the board. (where the Feedback pot is.)
From the feedback pot (P1) the signal works it's way to the switch pads.
The signal almost makes a 360 around the board!  

[Fender3D]

I saw it,
maybe feedback control was a last time addition...

The signal almost makes a 360 around the board!  

It's a phaser, after all...  

[armdnrdy]

I was planning to remedy the no feedback cap issue on my build with a couple of .1µf caps on either side of the feedback trimmer/control.

Nick...were you experiencing a loud pop when you engaged/disengaged the pedal?

[Fender3D]

I was planning to remedy the no feedback cap issue on my build with a couple of .1µf caps on either side of the feedback trimmer/control.

one cap should be enough.
You just need to keep DC away from output jack...

[armdnrdy]

Here's how I see the output signal/feedback path routing.

[nickbungus]

Larry, that all looks amazing.  Well done.

I didn't bother with the pedal switch from the schematic.  Its on my pcb board but I just took the output from Pedal Switch A (on the 1482 Schematic) and wired it as in the Output from the Beavis True bypass wiring diagram.  I've used this wiring quite a few times and you always get a pop, but not an unusually loud pop.  Please see the vids on pages 3 and 4 of this thread and can hear it.

[armdnrdy]

The pop doesn't seem that loud but....if you turn it up to stage volume....it will be! 

On your build, you can probably take care on that with a 1-10M pull down resistor from the board output to ground.

[nickbungus]

Especially as loud as I play.  The rest of them are constantly complaining!

[anotherjim]

Wow! I go off to eat my meatballs and pasta and there's incredible updates when I come back!
Terrific reverse engineering from Larry.

Oh, I failed to notice that DC on output thing... yeh, one cap , maybe between feedback trimmer and pot should do it  + anti-pop pull-down.

Input impedance down to 100k? Single coils only then  

I'd be tempted to put it between a TL072 for input buffer and output with some modest gain and vol control.

I've been wondering why all of the LM324 amps in the audio path have 10k pull-downs on the outputs? Something to make the amps more linear? Not something I've noticed in the data sheet.

[armdnrdy]

Wow! I go off to eat my meatballs and pasta and there's incredible updates when I come back!
Terrific reverse engineering from Larry.

What I lack in knowledge of electronics theory, (just about everything!) I make up for in electronics forensics and perseverance! 

I've been wondering why all of the LM324 amps in the audio path have 10k pull-downs on the outputs?

I too thought that this was a bit odd. I looked through a folder of various phaser reference circuits but couldn't find anything that used pull downs at each output of the phase stages.
This phaser design is pretty basic...so....I wonder if the pull downs are to keep the overall phased signal down to keep distortion at bay.

[armdnrdy]

I took another look at the 100K input resistor.

R1 is the only resistor that I can't see clearly on the board shots. It is covered by a large .1µf cap!

There is one image where I can make out the last three bands. (Black, Yellow, Gold)

The immediate area contains 100K resistors so...maybe that influenced my choice of the first band.

Since a 100K input resistor will give a gain of 1... and the original 270K a gain of .37 there is quite a difference.

So...I started thinking...what would be an obvious choice for R1?

If you look at R13 on the redrawn schematic, you'll see that the original 330K was changed to a 300K. Not much difference there.

To keep costs down...why source two different value resistors? Why not make the input 300K instead of 270K? Once again...not much difference there. (Gain of .33)

300K color code Orange, Black, Yellow, Gold.

Is orange the missing first band? 

[anotherjim]

Well, I don't know. 100k is possible. A well regarded British solid state amp from the 70's - the H||H IC100 had 100k inputs. Didn't worry Marc Bolan or Wilko Johnson to name 2 users (me too!).

Larry could be right about those amp pull-downs. Some op-amps produce a phase shift as they start to run out of bandwidth and the 324 has low bandwidth. Maybe the 10k load keeps it in check?

Have done some "mods" to Nicks MkI. Nick has fixed the board down, so I've found it easier to work from the top side. I kept the 2.2M R34 and fitted a 4.7M in the toe switch wire, so total = 6.9M. I like that. The bias control is easier to adjust and there is some tonal choice you can make as to where you centre the sweep. I can hear sweeping throughout the LFO cycle now. Of course, this is with R25 & C12 in place. Modulation is more than deep enough at fast rates, but maybe too deep for some.

Thinking a bit more about it, a modulation depth control which does not change the FET bias, could be had by manipulating R25.
C12 AC couples the divider so changing the ratio won't shift the DC level. The filter frequency formed with C12 would change with different resistors, but that might not be too bad.

I put a 100n in the wire to the feedback pot and tacked a 100k pull down between R51 and the ground end of R52. There was bypass popping before - there isn't now  That 100n value doesn't seem to alter the character of the resonance with high feedback - which can be nice and swampy.

[nickbungus]

Jim, please feel free to release the board if you want to do anything.  Its only held in place with a few sticky cable tie holders that I borrowed on a long term basis from work.  

Foxxy Number 1 seems to be enjoying his holiday to the valleys.  These mods have got me excited, I can't wait to hear it!  Although no rush to return it, keep it as long as you like.  You and Larry seem to be bouncing off each other and I believe Larry's is going to be amazing.

[armdnrdy]

You and Larry seem to be bouncing off each other......

Yeah...Jim's holding the steering wheel...and I've got my foot on the accelerator!  

Jim,

You might try lifting the R21 side of R25. I have a feeling that it will open the depth more. If it sounds too "over the top," the feedback control can be dialed back a bit.

On the subject of depth...It seems that a pot to ground at the LFO output would do the trick for a depth control.
Maybe even replace R25 with a pot....not sure what effect changing the resistance between the resistor and capacitor of the RC filter would have. Worth a try?

Edit:
I was looking at the schematic and had another thought. 

What if a 100K pot to ground were put at the output of IC3B? That should lower the amplitude going to the JFETs.
The pot could probably be connected anywhere after R43 as well if a "lower limit" resistor is connected between the pot and ground. Just a thought!