another EA Tremolo report: NOT working!!!

[keko]

I just finished my tremolo, but it ain't working as it should .

Built on perf, using runoffgroove's schem and layout I used 2N7000 for the MOSFET, 2n5457 for Q2 and 2N3904 for Q3.

At first it didn't even oscillate, althoug it did overdrive really nicely. I played over 30 minutes just drooling with that sound, till I remebered it was suposed to be a tremolo...  :?

Once I reversed the 3904 the led lit, and it tremoloed...(nice term!)

I have one problem, and one mod to solve.

Problem
The ckt works whenever it wants to. If I speed it up, it so fast that the effect is almost not noticeable. That's ok I guess.  BUT, when I speed it down again, it stops oscillating. Sometimes it takes a few second for it to start...but it also took it about a minute once. The recovery time is always different...

Mod
I didn't use a level pot, to make it simple at  first, and the ckt has so much gain it overdrives the clean channel on my amp (jfet pre, SS Trace Elliot).  Will this be solved using the Level control, or do I have to change some parts to reduce the overall gain? Wich ones?

Please help me...as most of you have already said, it's a great sounding effect, and I'd really like to understand how to get it working nicely (althoug, and I repeat, the distortion you get from it is awsome!!!)

[sunder]

mine uses a very similar oscillator, and it has the same phenomenon of needing some time to start oscillating. Set the speed pot at 50% and it'll start oscillating quickly. You need to find the right size speed pot - it should go fast to the point where it nearly gets too "blurry" and slow to the point where it almost stops oscillating. in other words, the working speed range of the oscillator should be layed out across the whole range of the pot. tinker with pot size.

A level pot should fix the other issue, I used a 100 k pot between output and ground and took the signal from the wiper. You can play with the value if you want.

[Boofhead]

> 2N3904 for Q3.

The gain of this transistor might be too low.  If you have a hfe tester, test a few and pick the one with the higher gain.  Alternatively use a higher gain transistor.

[Paul Marossy]

Two things come to mind:

1. Verify that the 2N7000 and BS170 have the same pinouts and that it is connected to the circuit properly. (I think they have the same pinouts, but I don't remember for sure off the top of my head)

2. The 2N3904 only has about 1/3 the gain of a 2N5088. This could also definitely be part of the problem with it not oscillating properly.

[smoguzbenjamin]

I used a BC 547C for my Q3, and that has only just enough gain to make the LFO oscillate. If you hang on a sec I'll dig up the board and measure the transistor Hfe.

[smoguzbenjamin]

Measured it, it has an Hfe of 597. Works for me. The level control is something you have to have, or you'll overdrive your amp, as you allready said. It's in there for a reason you know

[gez]

keko, if your cirucit is on a breadboard you might want to give this arrangement a go:



Sometimes doing things like this introduces noise, othertimes it's ok.  Haven't tried it with the EA, but I'd be interested to hear if it works out.

The (slight) advantage of doing it this way is that can attenuate the signal at the input which helps reduce signal distortion/avoids clipping the output if you feed it a large signal.  

Probably not that relevant here, but just thought I'd mention it.

[keko]

Gez, I'm gonna try that and let you know.

Thanks everybody for the quick answers.
But, regarding the gain this thing has....putting a pot in the output would just act like turning the volume down on your guitar, isn't it right?

Can't we just lower the gain at the input buffer? Or has it has to be high to drive the osci?

[smoguzbenjamin]

The whole point is that the gain of the MOSFET is changing constantly. When you turn the rate and depth all the way down, you can use it as a clean booster, that's the idea.

[gez]

Gez, I'm gonna try that and let you know.

Thanks keko.  If it's noisy in any way, just use the volume pot set-up as shown in the original runoff schematic (but you do need some sort of control to attenuate the signal).

[nirvanas silence]

The problem with oscillation stopping was prominent in my build as well.  Ended up being my battery was too low.  The few changes I made had created an even more sensitive circuit.  AC is the way to go.  I ran it at 18 volts too, fried the 2N5457 in a few seconds though.

[Boofhead]

Ended up being my battery was too low.

Raising the supply is actually hiding the real problem - not enough transistor gain.  

With the LED version, the circuit is more sensitive to the transistor gain - it needs to be reasonably high.

[keko]

I tryed several 2N3904, BC239 and 2N5457 for Q3. Now I'm using the BC239 with an hfe about 610, and the oscillation problem went again. I realizaed that lower gain also affected the rate's range, i.e. lower gain means lower maximum tremolo.

I think I blew the 2N7000, so I replaced it and it works fine. I can't get a 'no tremolo' sound yet. I figure I have to fiddle with the pot's values, or add a few ohms. I'll experiment and report back.

A question about biasing: runoffgrove's schem stated that the trim pot should be set till there were 4.5v at Q1's Drain. I kept turning the trimpot, but only Gate voltages changed. D was always around 9.2 and G changed from 0 to 6. Anyway, I think that when Gate (not drain) was biased to 4.5 I got sound out of it. Is this a typo in the schem?

To measure I just tap ground with the black probe, and Drain or Gate with de red one. Is that right?

BTW, I am running it from a filtered and regulated PS.

gez, I haven't been albe to test your buffer yet. I will do it tomorrow.

[gez]

There's no misprint.  You've either got the MOSFET wired up wrongly or you're taking measurements from the wrong place.

Looking at the schematic, attach the red lead of your meter to the junction of the 4k7,the output cap and the MOSFET (the drain connection),  and attach the black lead to ground.  Adjust the wiper of the trim up from ground until you measure 4.5V - 5V (not critical).

[gez]

PS.  Technically it's not  a buffer, I labelled the gif. wrongly when I posted the schematic (slip of the brain).  The arrangement I've shown is an ALTERNATIVE to the amp shown on the runnoff schematic (not a supplement!)  

[keko]

yep..shame shame shame on me.

When I was measuring the voltages, I had no transistors installed. I gave it a thought, and it was clearer...voltage interacts with the parts chosen..so now it works fine, at 4.7 v.

Also, once the volume pot was in there, the oscillation range varied a little bit. Now it works also as it should: a booster.

Gez, I have been really busy these last days, but I'll keep my word and try your schem out.