Improved EA Tremolo

[pauldumbell]

http://www.generalguitargadgets.com/projects/26-tremolo/142-ea-tremolo

Alright, so Im on my second build, and I assume that I will be a pro by now, right? WRONG. This pedal is a pain in the butt. First lesson. When ironing my transfers to my copper board, keep the steam on, which was a happy accident from my first pedal build. It took me 6 ruined transfers to realize that maybe the steam helped on this one. Lesson 2. Test all components, even if I just opened the package from mouser. I got a bunk transistor which caused me to waste a full day. Lesson 3.  Temperature control while soldering a pcb. Ive used the same iron forever, Its hot. REAL hot. Like burn the copper up off of the pcb hot if you hit the spot more than twice. Ordered a weller with temp control. BINGO. Solder points are much cleaner, and no nasty bridges.
So I built this pedal 2 times.
I had to buy a mpf 102 from radio crap since my n5457 was bunk

now I have oscillation, but its not on the signal, only the buzzing from a bad ground is tremoloing ....my aching skull.

I should go back to doing tube amps. Everything is BIG and easy. Nano world is about to give me an aneurism.
might this have something to do with the mpf102?

thanks all in advance!
pAul

[chptunes]

I don't have all of your answers, but I do have one note of input here.  In most JFET applications, the MPF102/2N5457/J201 are interchangeable.. but, I have read several times that the 2N5457 is much preferred over other FETs in this circuit.

MPF102's do provide a lot more gain than the 2N5457.

-Corey

[midwayfair]

Okay, well, let's see how much of the pedal is working first. If you pull the FET, do you have dry signal?

Sounds like your oscillator is working, just something wrong with the dry path, which is good, since the dry path is the easy part of this one to debug.

Double check your pinouts, too.

[Mark Hammer]

Apart from solder bridges and other sources of unwanted, unfulfilled, or unplanned connections, I think you'll find that JFET pinout will pose some aggravation.  Some manufacturers have datasheets with beautiful 3-d drawings of the transistor that identify the pins so straightforwardly, a 2 year-old could do it.  Others, not so much, and we don't always order/buy the JFET from the manufacturer with the most understandable datasheets.

I wish it was as easy to identify drain-source-gate on a JFET with one's meter as it is to identify emitter-collector-base.  

Corey, the JFETs used in the EA trem are used in two different ways.  What the GGG schematic shows as Q4 is used as a buffer/gain stage, though precious little gain is applied.  Q2, on the other hand, is used simply as a voltage-controlled resistor.  Drain and source can be flipped on that one with no impact on functioning.  On the other hand, if the drain pin is sitting where the gate is supposed to be, you're not gonna get much tremolo.

[duck_arse]

the mpf102 may have a higher than usable Vgsoff, according to a datasheet I was looking at today. the 2n5457 is a safer bet, and it's gate is the same place as the collector on a 2n5088, for example.

[pauldumbell]

So , I have the 5457 in q4 and the mpf102 in q2, do you think swapping places would be of use?then the mpf102 is the buffer.

[Mark Hammer]

I tend to see the 5457 used as a voltage-controlled resistor more often than an MPF102.  I do see the 102 used that way as well, just the 5457 a little more often.  So yeah, maybe you might have some luck with the two JFETs swapping places.  If you are using sockets, tin the ends of te trannies a bit so they make a good secure connection.

[pauldumbell]

thanks mark, Ill give that a shot. hopefully my next post will be, "thanks everything is as it should be."
by the way, this community is downright superb!
glad that I found my way here.
Paul

[pauldumbell]

swapped the transistors on q4 and q2, no noticeable change. tremolo is still very weak. i tried bringing the 560k resistor down to 470k then down to 320k and still no real difference. the only thing that i wonder about now is my volume pot, I used a 50k pot (because i had tons of em from my hill datum mixer) and chucked a 47k resistor across the outer lugs, but that shouldnt cause an issue.offboard wiring double and triple checked. Im stumped.

[duck_arse]

so, now is the time for you to give us voltages. the vol pot should be the least of your troubles. I'd try r7 or r9, as one goes up, the other goes down, and vice-versie. are you sure the osc works, and tracks the speed pot?

and why does the circuit have c9 in series with c1? remove and link c9, I'd say. for such a simple circuit, been around about 50 years, it's amazing just how much trouble this circuit causes. well, not the circuit itself ....

[pauldumbell]

yea, my voltages on q1 and q4 are all wrong.

q1
C 2.2
B 0.7
E 1.9

q4
Drain 9.5
Source 6.5
Gate 0.72

pulled c9 and no noticeable difference, so i put it back in

[pauldumbell]

seriously considering scrapping this ggg version and doing the tonepad version. after the motto less is more.

[duck_arse]

I'd check your pin-out for q1. again.

capacitors in series: the total value can never be more than the lowest value, so the 10uF is wasted here.

[GGBB]

seriously considering scrapping this ggg version and doing the tonepad version. after the motto less is more.

The homewrecker/runoffgroove version is both less and more.  I built that one and it sounds great.  http://www.home-wrecker.com/eatremolo.html.

[Paul Marossy]

I'm still perfectly happy with the original version of the EA Tremolo.

[pauldumbell]

So ladies and gents,

scrapped the GGG version and did the tonepad version. Big surprise, same problems. I had a massive meltdown after hours of measuring and troubleshooting my third attempt at building this pedal. I did what any sane man would do and ripped it apart and smashed the rest. I threw it so hard on the workshop floor that the hammond case is bent. Stuff was flying everywhere. The dog even hid in her cage. I think I even let out a fury of curse words, or was I speaking in tongues? Then I realized what the problem might be. 2n3904. If it has a low hfe, which it does...it might not be getting enough juice out to get the oscillator going. Of course I will have trouble checking my theory as all my pots for this project are destroyed by my own temperamental meat hooks. Oh curse thee EA tremolo, thou hath bested me.

[bluesdevil]

Real sorry to see ya giving up!! After a few successful builds under your belt, revisit it and go with the homewrecker version.... I know it works, I built it.
Hang in there and good luck!

[GGBB]

Bummer.  I hope you do get it going at some point in time.  For what it's worth, I don't think that the 2N3904's hfe is the problem as it is not part of the oscillator.  My understanding is that all Q1 does is control the gain applied to the audio signal.  Q2 acts as a variable resistor that is controlled by the oscillator Q3, and this variable resistance is used to change the AC gain of Q1.  And Q4 is just an input buffer.

[pauldumbell]

Ill give the runoffgroove layout a try. If I cant get that one to work, Ill move on to a superfuzz, which in its own right will probably cause me more misery.

[brett]

Hi
agree that Q1 voltages are all wrong, but Q4 isn't necessarily wrong. JFETs have lower voltage on the gate than source, and the 5457 needs a large difference to 'turn on'.
Q1 has something seriously wrong with the base voltage. It should be a volt and a half (wild guess). Without Q1 working you're not going to get the circuit working. The base should have a higher resistance to the power supply (560k) than to ground (150k). The exact values aren't critical, but they should be in that ballpark (ie 470k and 120k would probably work, but 560k and 15k wouldn't). I'm particularly suspicious of the base-ground resistor.
Good luck. This tremolo is really good.