Neovibe LFO (Sorry for flogging the dead horse)

[johnadon]

I hope I'm not posting prematurely, especially since there is a considerable amount of info in the forum and elsewhere on the topic of the Univibe LFO. RG's layout even when done on perfboard fits together like magic and I know my  Neovibe will be wonderful – once I get the LFO working properly. :-) I should mention that every problem I've ever had with a build turned out to be a wiring error, but I've never had quite as much trouble locating errors as in this one.

My problem: No oscillation in the LFO section. The audio portion works fine all the way up to and including the phasing induced by the LDR's. I flashed a flash light on and off at the LDRs and the phasing is quite audible.  I don't get any signal when the switch is set to Vibrato but I think that's another minor wiring issue in the signal path that I'll be able to fix. My big problem is that the LFO bulb remains lit all the time although it responds to the trimmer control all the way to darkness.

I've traced the LFO portion twice (actually the whole circuit) with multimeter/schematic, fixed the dumb wiring errors I found,  and checked orientations of caps and transistors with the layout and datasheets. My voltages all match the ones listed on RG's schematic.

I'm left with a couple of questions:

Should I be suspecting LFO transistors (I used all 2n3904's for the entire build) as the culprit? It sounds as if that's frequently the problem. I can post voltages on all legs after work if that would help. (Yes, I'm goofing off, but this is IMPORTANT! Haha!)

OR:

I made a parts substitution that may be causing the problem. I was very confused by a statement in the notes below the parts list in RG's neovibe project PDF, and I've used a regular non-stereo log pot for the speed control. He seemed to suggest using just a regular 100k log pot as a substitution, but also specifically directs the user to use the part listed in the parts list with is the standard 100k reverse audio log pot. I used a regular single log pot. Could this be the problem?

[R.G.]

Read this one: http://www.diystompboxes.com/smfforum/index.php?topic=69452.0

The 2N3904 works many times for the LFO transistors. Sometimes not. Also, the back to back diodes may cause you problems. The LFO is more delicate than I thought for years.

The simplest thing to do is to put a high gain transistor in for both LFO transistors; I like 2N5088, but 5089 and MPSA18 are even higher gain.

I have simulated using a JFET for the first transistor, and it seems to work. You can use a 2N5485, probably a J201 if you have them.

Some day I'll plow through the math to find out exactly the sensitive part or parts. Meanwhile, a few parts subs wil probably get you going.

The speed pot is fine. The exact taper or direction doesn't matter so much as the absolute resistance of the pot in terms of just getting it to oscillate. The taper and direction do matter when you put it into a rocker or panel.

By the way, riding (as opposed to flogging) a dead horse is a favorite topic of mine. Do a search on "dead horse" authored by me.

[johnadon]

Thanks, RG! Quite an interesting thread which I somehow missed. I've got all of those parts on hand and can't wait to get back up on the horse tonight when I get off work. I guess I'll be riding side-saddle.

Also, thanks for all your work in general and especially the insights you provide on these forums.

John

[johnadon]

2N5088's = Magic!
R.G. = Magic!

The Neovibe's working perfectly, and now I've got a hankering to play around with the JFET trick you mentioned at some point.

[jball1985]

How can i fit a tip112 in the driver spot this thing is huge!

[jball1985]

disreguard my post i feel stupid