Phase 100 rant, and question: TL022 = TL072 ?

[Praying_V]

Hey everyone.  I've been working on two Phase 100's (for way too long now).  If you ever want a reliable, good sounding phaser project, then avoid this one like your life depends on it (search the forum, you'll see how many people were unhappy with the Phase 100s they built, or tried to build.)

The first one I built worked & sounded great- for about ten minutes.  I'll never know what happened after those ten minutes!  It was sounding GREAT, then all of a sudden, CRAPPY.  Its still phasing, but its weak & puny.  At some switch settings, the phase is inaudible.  LFO is working perfectly.  Voltages look good.  I even used matched vactrols too... then WHY???

The second one I'm working on is a vintage unit, with two fried vactrols.  How does a vactrol end up getting destroyed anyway?  It will hopefully work fine once I get new vactrols in it (but who knows).

A forum member mentioned in an old post that he had an old MXR factory set-up sheet, and that setting up a Phase 100 wasn't as simple at tweaking a trimpot.  He said there were several voltages which all needed to be adjusted on each unit.  He didn't get any more specific than that-  I've PM'd him but can't get a response... 

...Getting to my question:  The vintage unit & schematic uses TL022s for all the opamps.  The project file, available at GGG, uses TL072s.  I used TL072s.  Is it reasonable to think that using the original opamps (TL022s) could make all the difference here?  I'm getting desperate...  I'm open to any other suggestions as well!

Schematic-- http://www.generalguitargadgets.com/pdf/ggg_p100_sc.pdf
project file-- http://www.generalguitargadgets.com/pdf/ggg_p100_lo.pdf

Thanks ..............................ken

[B Tremblay]

I don't have any answers, but can offer company in your misery.  A couple years back, I used the GGG layout to build one and never got it to operate correctly.  I reflowed all the connections then checked every component, wire, and trace to no avail.  In the end, I decided to shelve it.

[DougH]

The Phase100 is one of those effects that goes in the "for 130 bucks I'll buy one" category for me. My time's worth more than that.

[Ry]

I've got one that Bill Bergman built.  It sounds great.  I don't know what layout he used, but if you need voltage references, I can provide them.

[mdh]

I don't have any direct experience with this circuit, but I would be very surprised if the op-amp choices were that critical.  TL022s usually come up in LFOs, where low current draw can help prevent ticking, but most of the op-amps in that circuit are just plain vanilla phase shift stages like you'd see in most any phaser.

As for blowing vactrols, it seems the surest way is to allow the LED to draw too much current.  So maybe in troubleshooting the one with the bad vactrols, you should first install some sacrificial LEDs and see how much current they draw when the LFO is going at maximum depth.  If that's close to the maximum spec for the LED half of the vactrol, you'll be one step closer to identifying the problem.

[Praying_V]

Ry-- That would be GREAT, let me post tomorrow when I've had a chance to get the project back out & see what my voltages look like.  Believe me, it would be MUCH appreciated, and might help others too-  there's more than a few people on this forum over the years who gave up on this one.  There's not enough info available right now...  I know these pedals should sound great.  You'll be a hero!

B. Tremblay- I hear ya!  I shelved mine too, both of them...  But its been eating away at me, so I'm ready for round two.  Wish me luck...

[B Tremblay]

B. Tremblay- I hear ya!  I shelved mine too, both of them...  But its been eating away at me, so I'm ready for round two.  Wish me luck...

I wish you luck and more importantly, patience!

[slideman82]

You could try many dual op amps, try TL062, draws less current than TL072. Just put a socket, and start trying many dual op amps, LM358, 4558D... and list continues to infinity!

[Ry]

I don't know about being a hero, but I will certainly try to help you out  .  I'm not too familiar with the circuit itself, so it may take me a bit to relate the schematic to the layout that Bill used for some voltages.

Let me know what you need and I should be able to get to it on Friday (my next day off).

[soulsonic]

I haven't tried building a Phase 100, but I can say from experience repairing other phasers that weak crappy phasing is caused by incorrectly functioning variable resistance in the phase stages - ie; the vactrols aren't working right.

If the LFO is indeed blowing up the vactrols, then increasing the value of R54 (in the GGG scheme) would be the first place I'd start looking. Maybe the value is too low for the kind of transistors used for Q1 and Q2? Did you use 2N5088? I wonder how the "On" resistance of those compares to that of the 2SC1849?

Should all the LEDs be wired in Series? That's how it is shown in the schematic. I wouldn't do it that way. That could be another source for this problem.

[Praying_V]

OK, I've measured all my voltages... and I mean ALL of them.  There's 4 transistors, 6 dual-opamps, 3 vactrols, and a diode.  I'm using the part # designations from the GGG layout (link to the layout & schematic are in my above post ^).  All the opamps have roughly the same steady pin voltages, except for pins 5, 6, 7 on IC1 which make up the oscillator.  For the vactrols I only measured the voltages on one the LED pins, which show that the oscillator is working properly.  The pin voltages on the LEDs, IC1b, and most of the transistor pins all change with the LFO, so measuring them can be difficult.  Its easier if you turn the speed down to the lowest setting.  I recorded the highest & lowest points of their voltage ranges. 

One last thing worth mentioning- I'm using a Boss adapter @ 9.7V (could this be a problem? I doubt it...).

IC1
1: 4.88      8: 9.73
2: 4.83      7: 1.28 - 9.00 (square wave)
3: 4.83      6: 3.7   - 6.0   (smooth)
4: 0           5: 3.57 - 6.18 (square wave)

IC2
1: 4.87      8: 9.73
2: 4.82      7: 4.82
3: 4.83      6: 4.82
4: 0           5: 4.80

IC3
1: 4.89      8: 9.73
2: 4.88      7: 4.82
3: 4.86      6: 4.87
4: 0           5: 4.85

IC4
1: 4.89      8: 9.73
2: 4.88      7: 4.87
3: 4.42      6: 4.87
4: 0           5: 4.85

IC5
1: 4.88      8: 9.73
2: 4.88      7: 4.86
3: 4.86      6: 4.87
4: 0           5: 4.84

IC6
1: 4.90      8: 9.73
2: 4.88      7: 4.86
3: 4.86      6: 4.87
4: 0           5: 4.85

VACTROL LEDs: These are wired in series.  As the LFO cycles, the voltage drop of each LED moves up & down. Voltages shown are measured to ground, from the anode & cathode of each vactrol.

OC3
(+) 9.73
(-)  8.18 - 8.27

OC2
(+) 8.18 - 8.27
(-)  6.65 - 6.83

OC1
(+) 6.65 - 6.83
(-)  5.11 - 5.37

TRANSISTORS: Voltages measured to ground.  Although small, there were noticeable cyclic fluctuations on most of their pins, so I recorded the highest & lowest voltages of their sweeps:

Q1
C: 5.09 - 5.35
B: 0.53 - 0.61
E: .003 - .020

Q2
C: 5.09 - 5.35
B: 0.90 - 1.04
E: 0.53 - 0.61

Q3
C: .755 - .760
B: .586
E: 0

Q4
C: 9.63
B: 3.8 - 6.2
E: 3.3 - 5.6

[Ry]

I don't think your AC adapter has anything to do with your problem.

Here are my results with a fresh battery measuring 9.45 V.  They match up pretty well with yours.

The ICs are LM1458's and the transistors are BC107b types

IC1
1: 4.63      8: 9.45
2: 4.71      7: 2.1 - 8.8
3: 4.70      6: 4.0 - 6.0   
4: 0           5: 3.8 - 6.0

IC2
1: 4.76      8: 9.45
2: 4.70      7: 4.70
3: 4.7        6: 4.70
4: 0           5: 4.70

IC3
1: 4.60      8: 9.45
2: 4.70      7: 4.60
3: 4.70      6: 4.70
4: 0           5: 4.70

IC4
1: 4.60      8: 9.45
2: 4.60      7: 4.73
3: 4.30      6: 4.68
4: 0           5: 4.70

IC5
1: 4.63      8: 9.45
2: 4.70      7: 4.80
3: 4.63      6: 4.70
4: 0           5: 4.67

IC6
1: 4.60      8: 9.45
2: 4.70      7: 4.75
3: 4.60      6: 4.70
4: 0           5: 4.60

Vactrols:
OC3
(+) 9.45
(-)  7.89 - 7.92

OC2
(+) 7.89 - 7.92
(-)  6.39 - 6.46

OC1
(+) 6.40 - 6.845
(-)  4.9 - 5.0

Transistors:
Q1
C: 4.87 - 4.96
B: 0.621 - 0.637
E: .007 - .012

Q2
C: 4.85 - 4.94
B: 0.96 - 1.09
E: 0.598 - 0.63

Q3
C: .911 - .922
B: .607 - .609
E: 0

Q4
C: 9.37
B: 3.97 - 6.0
E: 3.3 - 5.43

[Praying_V]

Thanks for those voltages Ry!  You're right, ours both look pretty much the same.  Although there's one thing about Bill Bergman's circuit that catches my eye- the ICs.

I recently built a Nurse Quacky.  I used TL072 for the IC, which was recommended in an older project file, and by at least one person on this forum.  Well, the pedal DID quack, but it was very weak, and not sensitive enough to be considered usable.  After some more research, I found out that several other forumites had the same problem- only when using the TL072.  The cure?  Replace the TL072 with an LM1458!  When I did, the difference was like night & day, and it was a fantastic sounding filter.

I don't think its a coincidence that we're talking about the same two opamps, and very similar symptoms.  I'm ordering some TL1458s & making the switch.  I'll also compare the 'on'-resistance of the 2N5088 with the original 2SC1849 transistors, and also the BC107b that Bill Bergman used.  I'm not having problems with the vactrols blowing up (that's my OTHER Phase 100, a vintage unit, with the proper transistors), but since I have some BC107s here, I may as well see which transistor gets closer to the original's specs.

[B Tremblay]

This morning, I decided to pull my abandoned Phase 100 build out of the drawer to try some different op-amps.  I scrounged around until I found two LM358 and one 1458.  I'd previously tried TL072 and NE5532 in my build, but never got to the point where I felt I had a properly working unit.

First off, I did make some mods.  I used the GGG project, so my references are specific to those documents.  However, I started the build two years ago and JD has changed the layout slightly.  I don't think there are any changes to the component numbering, though.

I omitted the left side of the Intensity switch, the half that has one pole wired to the R41/R42 junction and the other pole wired to the R51/R45/Q2 base junction.  I like plenty of color in my phasers, so leaving that switch open sounded better to me.

I replaced the right side of the Intensity switch (that switches between the R21/R22 junction and R1) with a 10k pot wired as a variable resistor.

I changed R56 from 10k to 4k7 for higher speeds.  The other differences between the schematic and my build were dictated by what parts I had on hand.  I used a 2k2 for R42, normally 2k4.  I used 22u for C17 and C14 (15u on the schematic), and a 10u for C15 (15u on the schematic)

After spending a long time swapping ICs and trying in vain to observe the differences of the endless permutations, I turned my attention to the 20k trimpot that adjusts the center frequency.  Labeled as R47, I'd made it a panel pot for easier adjustment while testing.  I found that I needed to have the pot set to its minimum setting to have any high end to the sweep.  It was still very asymmetrical, so I decided to try a larger pot.  I incrementally increased the pot value and settled on a 100k-linear pot.  I now have much better control over the sweep symmetry.

So, I now have the same switch mods, a 100k-linear pot for R47, and all ICs are NE5532.

I'm the happiest I've been with the sound so far.  One thing I noticed is that while different ICs did affect the sound, using the 5x larger pot for R47 allowed me to adjust the center frequency enough to compensate for the difference in tone.

[spudulike]

I recently built a Nurse Quacky.  I used TL072 for the IC, which was recommended in an older project file, and by at least one person on this forum.

Full build and component swap report here ...
http://www.diystompboxes.com/smfforum/index.php?topic=35760.msg251999#msg251999
BTW I found the best combo to be NE5532 and BC549B.

[Praying_V]

OK, the Phase 100 is now working perfectly, and sounds really good too!  I switched all the ICs from TL072, to LM1458.  That's all it took!

I re-tested it with the TL072's just before making the swap.  The phasing was barely audible (if at all), although at one switch setting there was a loud whistling oscillation modulating with the LFO (described by others too).  Not cool.  So I pulled the ICs and put in six LM1458's.  Now there's a great full sounding phase, and at the extreme switch settings it only sounds fuller - with the ICs changed, the oscillation is gone.

I'd say that this circuit is un-build-able using the specified TL072 chips.  I haven't tried NE5532's, but it's great with LM1458's, and I can't even imagine it sounding any better than this.

[the recluse]

This is very interesting.  I bought and built the P100 kit from GGG and had some issues with it.  At one point I sent it back to JD to investigate because two of the switch positions resulted in insane oscillations.  I get an ok phasing on two of the switch positions, but the other two are just so so despite tweaking the trimmer a number of times.  Perhaps I'll order me some LM1458s and see what happens.

[petemoore]

  The only thing I've noticed with opamps is that LM741's sound a little different, they all sound may ever slightly different, but any of the duals worked mostly similarly except:
  In cases of ticking, where the TL062 might cure all or some of it, being that they draw less current, mostly noticed that with the EZ Vibe.
  I shelved mine too, both of them...  But its been eating away at me, so I'm ready for round two.
  There's a pattern developing ?
  Praying V seems to have energy to put into it.
  Could be there's a wire going the wrong way ?
  How different is the layout to the one that works ?
  If the schematic is true enough to the one that works, the ones with problems layout could be gone over a few times to see if it jibes with the known good schematic.
   I found that I needed to have the pot set to its minimum setting to have any high end to the sweep. Tremblay, perhaps having the total pot a higher resistance helps, but it would have been possible to add a resistor to the min side of the pot which'd put the center adjustment closer to center of pot and be a smaller change to the cross-pot resistance [from what I can tell, a little bigger value just adjusts differently, and doesn't really change the effect otherwise.
 

[the recluse]

I had 6 4558s lying around and tried those.  I don't think it made a difference from the TL072s.  Basically I really only get noticeable phasing on the 2 and 4 positions if I the rate knob maxed.  1 and 3 sound fine.

I'm curious, in order to stop the oscillations, JD bumped up the value of R1.  He actually noted this in the instructions for the kit and now includes an alternate resistor.  I wonder if I can change this to some middle value to bring out the sound on 2 and 4.

I'm still wondering about the 1458s though, might be time to place an order.

[Ry]

It's great to see so many people getting their P100's working  !

I had the same problem with my Bass Balls.  Once I put in a 1458, it worked perfectly.