Pulsar Debugging Help?

[Shandy]

Hi folks--

I've built a few (half a dozen) successful effects, mostly overdrive circuits, a couple compressors.  Right now I'm on day 3 of debugging my E.H. Pulsar build, but my feeble and shallow grasp of the actual principles of electronics in general can only get me so far.  Time to call in the pros.

Here's the lowdown:

EH Pulsar trem, schematic and layout from tonepad: http://www.tonepad.com/getFileInfo.asp?id=29

I do have one part substitution, and that is I used a .01uf cap in place of the .022uf cap for C7 until I can get my hands on the right one.

The only circuit modification is that I bypassed the AC adapter.

The problem: The effect passes the same clean, unmodified signal when the stomp switch is in either state.  The signal is nice and strong, just completely unaffected by the circuit-- no volume boost or attenuation, no distortion, no nothin'.

The out-of-circuit battery voltage is 9.1v

With an instrument cable plugged into the input and the battery in its snap, the reading at the end of the red lead is 9.06v

Now here's where it all goes wrong...

Q1
C= 0v
B= 0.55v
E= 0v

Q2
C= 7.19v
B= 0.51v
E= 0v

Q3
C= 3.45v
B= 0v
E= 0v

U1
pin 1= 4.79v
pin 2= 4.79v
pin 3= 4.55v
pin 4= 0v
pin 5= 0v
pin 6= 8.42v
pin 7= 8.42v
pin 8= 9.07v

Normal smart folks have probably spotted my issue pretty quickly, but I'm really having a hard time with this one.  I did check all of my pinouts on the transistors, the IC and I checked the polarities of the electrolytic caps, as well as the values of all the resistors.  I also checked continuity between every component in the build.  Obviously I overlooked something, but I'm just not finding it.

If you need it, let me know and I'll post a diagram of my off-board setup...

Thanks in advance...

[theehman]

Hard to tell without seeing your layoput, but I'd say the problem is in the oscillator section.  Have you audio probed the oscillator to see if it's functioning?

[oldrocker]

Also make sure that R8,R17,R18,R19 and R20 are connected together and no to ground.  Sometimes that gets misinterpreted.

[Shandy]

Thanks Ron... I just put together an audio probe.  Now [looks down sheepishly], in which area would I find the oscillator section?

Randy, I double checked those resistors, and they are where they ought to be.  But I can certainly see where people might look at it a little fast and assume they are going to ground.  Thanks...

[mdh]

Seems suspicious that the Q1 collector is sitting at 0V.  I would look around there for a short to ground.

[theehman]

Thanks Ron... I just put together an audio probe.  Now [looks down sheepishly], in which area would I find the oscillator section?

Check the collector of Q2 and see if you get a pulse there.  If so, go to Q1 and check the base there.

[Shandy]

Okay, I poked around all over and couldn't find anything going to ground that shouldn't be.  And there definitely seems to be an oscillator problem, as I'm not getting a pulse at Q2, or Q1 or anywhere else for that matter.  At least I know where to start though, right? 

[theehman]

Okay, I poked around all over and couldn't find anything going to ground that shouldn't be.  And there definitely seems to be an oscillator problem, as I'm not getting a pulse at Q2, or Q1 or anywhere else for that matter.  At least I know where to start though, right? 

That's where I'd start.  I think I'd take a closer look at Q3.

[Shandy]

Okay, here's what I've got...

I started in with my audio probe, beginning where the signal enters the circuit, and working my way from there.  I heard the guitar coming through loud and clear all through the schematic until I hit R5 (39k) as well as R6 (10k).  At each of those resistors, the signal disappears.  I measured them with my multimeter, and they check out fine.  Signal on one side, none on the other.

I know I'm overlooking something obvious, but I think my brain might be a little cooked on this project.  How can R5 and R6 be to spec but not allow a signal to pass?  Thanks for your help so far, by the way.   

[mdh]

I don't think R5 and R6 are in the audio path exactly.  I think that if you have audio signal at the output of the op-amp, then probing the audio path isn't gonna help you solve the problem.  As others have said, the problem is most likely the oscillator.  If you don't have a pulse on the collector of Q2, then there's something wrong in that portion of the circuit.  I still think it's odd that Q1 C is at ground potential, though.

[geertjacobs]

- Which type of transistors are you using?
Re-double check the pinouts, that solved all the problems in my pulsar build.
- is there 9V on your pcb on each location where there should be 9V? (no cracked traces on the pcb or so)
- is there 0V on your pcb on each location where there should be 0V?

PS check the pcb trace fix in the build reports on Tonepad.

[Shandy]

- Which type of transistors are you using?

2n5089...

Re-double check the pinouts, that solved all the problems in my pulsar build.

*Sigh* I almost didn't bother with this one, because I had taken so much time to look up the pinout online, print it out, look twice, etc.,.. and yep, I STILL got it wrong.     So, I pulled them and put them back in the PROPER way, but It seems to have had no effect, as the condition is exactly the same (clean signal, no pulses anywhere on the board).  Can I safely assume that I've toasted one or all of my transistors and need new ones?

If it matters, these are now my voltages:

Q1
C= 0v
B= 0.55v
E= 0v
(Same as before, actually)

Q2
C= 7.19v
B= 0.51v
E= 0v

Q3
C= 3.45v
B= 0v
E= 0v

The IC voltages are still the same:

U1
pin 1= 4.79v
pin 2= 4.79v
pin 3= 4.55v
pin 4= 0v
pin 5= 0v
pin 6= 8.42v
pin 7= 8.42v
pin 8= 9.07v

PS check the pcb trace fix in the build reports on Tonepad.

I did, and made the modification.  Thanks. 

[geertjacobs]

Quoted from R.G. Keen: explanation an how the pulsar works:
The Pulsar uses the variable resistance region of an NPN bipolar transistor as part of the gain controlling elements for an opamp.

The transistor is either turned more or less on by the trickle of current through that high resistance attached to its base.

The LFO is a discrete transistor astable oscillator, making a square wave at its output. The square wave is either fed semi-directly to the base resistor, or filtered to a more-or-less triangle or trapezoid by the selection of a capacitor on the base circuit.

Switching in a shunt capacitor to shunt across the transistor ups the opamp's gain to "bypass".

If your transistor is fried, it ain't gonna work.

As others said before: your clean path (straight through opamp) is fine, but the oscillating signal isn't getting to the opamp circuit.
So either your oscillator doesn't oscillate, or the path between opamp and oscillator is broken somewhere.

[Shandy]

Just an update, I replaced the transistors (I was out of 5089s, so I used 3904s) and I still have precisely the same voltages and the same tremolo-free performance.  So tomorrow morning I'll fire up the DMM again and look for a short or a break in the path from the oscillator section to the opamp...

[Shandy]

Alrighty...

I went over every inch of the circuit, and could not find anything shorting to ground, not connecting to the next component on the schem, or not going to 9v when it should be, and I specifically spent a lot of time around the collectors of Q1 and Q2.

One possible clue that I haven't yet mentioned is that when the effeect is engaged, R10 acts as a volume pot.  Maybe that's something?  I checked every offboard connection as well, and in the chance that I just don't have it hooked up as it should be, I've put together a diagram of my offboard setup, below.

I really appreciate all your suggestions and help, guys.  At this point though, I'm thinking that if I don't get anywhere by the end of the day I'll pack it in and salvage the usable parts for another project.  Hmm... maybe a chorus? 

[theehman]

I've actually got the prototype for the EH Pulsar.  I'll check the voltages on it tonite and compare them to yours. 

[Shandy]

Very cool of you to do so, Ron.  Thanks!

[theehman]

Here's what I got on my Pulsar prototype:

Q1
C= 0v
B=  changing negative voltage from the pulse
E= 0v

Q2
C= less than 1v changing w/ pulse
B= voltage swinging from + to -
E= 0v

Q3
C= changing positive voltage from the pulse
B= voltage swinging from + to -
E= 0v

And here's a pic of my 1979 EH Pulsar prototype:


This is the ACTUAL prototype.  It was purchased from an effects dealer who got it from a tech that worked in EH's UK office.  It works perfectly but doesn't have a footswitch.  You turn off the trem by shorting 2 wires together.

[Shandy]

Whoa, an actual working EH Prototype is pretty dern cool.  I have an Epiphone prototype solid body guitar that I like quite a bit, but it seems no one else did, as it never went into production.  It's pretty much just a flat-top double-cutaway LP, which I'm sure they've eventually produced at some point.

Well, my Q1 voltages sort of make sense compared to yours, but as far as the rest of it goes, there's some serious trouble with my oscillator section that I just can't iron out.  I never given up on a build before, but I think it's time to call it a day and get on with something else.  Thanks for all of your help, Dan...

[Shandy]

Problem solved!

Thanks to Ron (the EH man)-- he found a bad jumper in my wiring, and set it right.  It's sounding great now, and I'm really pleased with the effect... I recommend the build to anyone looking for a different take on tremolo, as long as you check your work more carefully than I did.