Vox - Repeat Percussion (V 809)

poodlebra · Post by **poodlebra** » 27 Jul 2010, 14:37

i'm happy (ecstatic maybe) to confirm that my vero works. i've no idea why the previous couple of attempts didn't. poor builds, i guess.

after reading through all the various suggested replacements on here for increased output and less ticking, i've gone with:

Q2 - MPSA18
R8 - 15k
R9 - 33k
R13 - 1.5k

C2 - DPDT (on-on) switch with 10u / 4.7u for slow/fast settings.

I moved the trace cut under the rate pot to between the other two lugs

rate pot works as an 'increased gap' rather than speed, i.e. it gets slower as you turn clockwise. suits me.

i found that adding a rate LED gave a more pronounced tick, so i've ditched that and sticking with my ears.

thanks for everyone's input. this is kick-ass.

acidfuzz · Post by **acidfuzz** » 09 Sep 2010, 04:35

Hey all,

I'm new to this forum + was just wondering if anyone had any tips on the 2n2646 uni-junction transistors.

I tried to build a Vox repeat percussion back in 2006 from a schematic I got from Ron Neely + had no luck. Neither he, nor I knew the schematic was wrong at the time. More recently I've built 2 based on the corrected schematic + both work + sound just like my original 60's repeater, however they have the limited rate/speed issue mentioned before. When I install the 2n2646 from my original vox repeat percussion unit, the circuits work perfectly without shutting off at higher rate. I bought 8 (eight) 2n2646's off a UK supplier on ebay + they all have slight variations in terms of how well they work in the circuit, however ALL 8 stick at high rate + I loose the oscillation. The only 2n2646 that works properly is the one from my original unit. Does anyone have any tips on 2n2646's ? i.e. Are there slight differences in specs between manufacturer/brand or year of manufacture? Has anyone found a way to determine if a certain 2n2646 will work BEFORE ordering them? I love this effect + would like to make more of these for friends + possibly to sell, but I don't want to buy so many just to get one that works properly.

Also, I've noticed that the circuit cuts a significant amount of bass freq. when engaged, which is cool in it's own way. Has anyone experimented with changing cap values to get a bit more full freq. through it?

Thanks

poodlebra · Post by **poodlebra** » 10 Sep 2010, 12:31

i don't know enough to comment on the UJT stuff, but mine doesn't stick open at all, even with the switch added for slow/fast range giving much higher speeds. can't remember off hand, but i think it's probably to do with one of the changed resistor values above. great pedal. love it.

theehman · Post by **theehman** » 10 Sep 2010, 12:37

I've built and sold about 50 of these and I've never seen any "stick" at maximum Rate settings. Is it just getting so fast it can't modulate the switching transistor quick enough to turn it on?

acidfuzz · Post by **acidfuzz** » 11 Sep 2010, 03:15

I'm not exactly sure why the new UJT's stop modulating. I've built 2 versions of the circuit with slightly different layouts + both seem to perform + sound the same. If I install the 2n2646 from the original vox unit I have on hand, both circuits function perfectly + sound just like the vox unit in all respects. I have 2 builds from your corrected schematic + both work perfectly with the og UJT, however with all of the "new" 2n2646's installed, the modulation stops after approx. 3/4 rotation of the 50K linear pot I'm using. I haven't changed the 10uf cap values to bump up the rate, + I used a 5k6 resistor between the UJT's emitter + the pot. I see that others have had the same/similar problem + I was wondering if anyone's had better performance from certain make/brand or year of make 2n2646's in this circuit.

theehman · Post by **theehman** » 11 Sep 2010, 12:12

Most of my selected ones are Motorolas. There's a few GE units in there as well.

R.G. · Post by **R.G.** » 12 Sep 2010, 02:35

UJTs have quirks. They're not as variable as, say, JFETs, but they do vary quite a bit.

The UJT oscillator depends on getting enough voltage to turn on the base junction in spite of the resistive drop down the emitter to emitter channel, and then on getting both low enough in voltage and current to stop the channel conducting.

If you feed the timing capacitor a lot of current through the timing resistor, you can get into a condition where there is enough current to continuously feed the channel and it doesn't switch off. I suspect that this is what's happening when it "sticks" at fast speeds. The solution is to make the timing capacitor smaller and the timing resistor bigger.

I've built a couple of PUT-based units now, and they don't have this problem as long as you don't artificially prop up the cathode side with a resistor that's too big. It may work the same way with UJTs, but I haven't tried it.

The PUT was developed precisely because of the variability of the UJT.

dieseat · Post by **dieseat** » 27 Sep 2010, 00:15

Recently, while ebay-browsing for parts, I came across the unijunction trannies and thought, "cool, but what are those good for again?" ...googled... got vox repeat percussion, ordered the parts (which were in fact PUTs), dug out my spacemen3 cassettes, waited, received, breadboarded, adjusted, voila! tick...tick...tick.

It did indeed modulate the volume, but the ticking was sooooo loud - I mean, way above line level, so that it sounded as if I was throwing a house-techno party through a pitchshifter. I read through various threads, switched between 2n5088, 2n5089, mpsa18, tried out different resistance variations around transistors with no success. I then looked for a possible defective cap and as soon as I removed the 100uF cap... no more clicking, yet still modulating, sounding excellent. Nonetheless, I am sort of wary about going on stage to show off my new home-made retro gadget and the thing dies because I omitted a component.

...finally to the point: what's the 100uF cap for? i.e. why is my breadboarded working fantastically without it. I originally thought it was so that the discharge doesn't sound too square or short. Without it, it doesn't sound that way at all. maybe a UJT vs. PUT thing? Also, I'm using 2n6028s which seem to have much lower current...

2. I'm getting the "sticking" as well, but it seems to vary some with the rate cap in relation to the resistance node around the gate of the PUT, so getting the right ratio may be the key to getting rid of that.

disclaimer: I'm new here and aside from maintenance and part replacement, I'm also relatively new to electronics, so excuse any dumb questions.

dieseat · Post by **dieseat** » 27 Sep 2010, 08:45

Oh... looking at the old schem, it seems it should be 100n or p, since a 100uF bipolar cap, or a 100F cap seems unlikely.

In fact, a lot of those component values are different, which is partly covered in the thread.

jdanner723 · Post by **jdanner723** » 11 Nov 2010, 14:03

Hi everyone! This is my first post but I've been using the resources on the website here for quite awhile. I actually got this schematic from here a few months ago and have been working on this project dilligently since. I guess I should start by saying that this is my first build so I'm still trying to get up to speed with some of the lingo and techniques.

I decided to throw this guy together on breadboard before soldering to my final perfboard. As of right now I'm pretty convinced that all of the components are in the right place and are properly connected. My main problem is trying to determine the wiring of the components into the breadboard. I decided for this part of the construction to skip wiring the switch so that the circuit is on continuously. I figured I can mess with the LED, power, and switching once I've determined whether or not my components will work together. Is this a good idea? Or is it better to go ahead and wire everything together even if I am using breadboard?

My second question is in regards to the wiring of the pots. I have what I think is the right wiring from the lugs to the breadboard but I feel like I'm missing something. Do any of the lugs need to be connected to the input/output jacks? Can anyone explain the proper way to wire the lugs to the rest of my project? I'm a little confused on this. I saw a layout for stripboard that had the 10K pot being wired to the output but I don't see any sort of indication of that on the schematic. Additionally, I'm a little lost with the wiring of the input/output jacks to the breadboard. I'm using switchcraft jacks #12A (output) and #12B (input). I don't know if these are the proper jacks to use but does it make that much of a difference? The output part of the schematic seems to have 3 wires running for the jack to the board anyway so I was thinking this one was okay to use.

Finally, and I think this is my last question, the majority of schematics I've seen show the 9V battery adapter connected to the input jack, at least that's what I've seen over and over on the Beavis Audio Research site, however this schematic shows that it's attached to the output jack. Is there anything special I need to know about this?

I apologize for not being able to provide pictures of my project currrently as I'm sure this would help out. If I can get my hands on a camera for a few minutes I'll try to get some shots.

Thank you in advance for your help and guidance. Sorry to write a novel for my first post!

J.

fifthdimension · Post by **fifthdimension** » 15 Dec 2010, 12:22

Hello,
I've an old Vox guitar with built in effects.

The only problem is that the repeat percussion isn't working anymore (or a little).
When I put the volume of my amp at 10, I can heard the sound of the effect, and I hear breath and buzzing.
There are 3 transistors (2 BC109B and 1 2N2646), some capacitor (0.068, 0.047, 10 bipolar, 100) and some resistors.

I think some components are dead, maybe just one capacitor and one transistor, I don't know.
Can anybody help me, what components can I check with a multimeter?

Thanks

fifthdimension · Post by **fifthdimension** » 16 Dec 2010, 12:02

I've tested all of the capacitors, apart the 100 because my mulimeter can't mesure high capacitor.
They all worked well.

RnFR · Post by **RnFR** » 17 Dec 2010, 00:41

fifthdimension wrote:I've tested all of the capacitors, apart the 100 because my mulimeter can't mesure high capacitor.
They all worked well.

why are you posting the same thing in two threads? please use the other thread for debugging. this one is for circuit analysis.

acidfuzz · Post by **acidfuzz** » 26 Dec 2010, 08:59

FInally had some time to build more repeat percussion units over winter break. Noticed this circuit is a very picky about which 2n2646's you use. Brand of manufacture doesn't appear to matter. Also noticed changing the timing resistor + cap helped a bit, as R.G. suggested. However, my early builds from the schematic floating around were all a bit pickier than the OG VoxRPC. So, completely disassembled an original unit + measured all components. One thing that made a huge difference, is changing the rate pot to the actual value used in the original unit, a 100K reverse audio taper pot. Incidentally, the schematic is still wrong in this respect + the rate pot should be 100K reverse audio taper. The rate pot from my OG Vox V809 plug in unit actually measured 130K! The rate pot in my Vox Ultrasonic guitar was ~100K.

Made a PCB that fits nicely in to most enclosures + incorporates the depth mod, which is quite useful. Also made versions with the timing cap switch, however I found the faster rate to be a bit cheesy IMO. Could be good for modulating up from a slower rate in a live situation like Spacemen 3 used to do. But having to interrupt the rate transition by flipping a switch to change capacitors makes the much faster rate useless for my playing. THe stock 10uf cap offers a nice wide variation of rate if used with a 100K reverse audio pot.

To answer jdanner723's question: "Finally, and I think this is my last question, the majority of schematics I've seen show the 9V battery adapter connected to the input jack, at least that's what I've seen over and over on the Beavis Audio Research site, however this schematic shows that it's attached to the output jack. Is there anything special I need to know about this?"

Nothing special here. Essentially what's going on is that the schematic shown on this site is from the original Vox/Thomas Organ schematic for the plug in V809 unit which had a MALE mono 1/4" instrument plug for the "input". The idea was that these units would plug directly in to your guitars output via this plug + you would then run your cable from the V809's output to your amp. So, they used a stereo output jack to do the battery switching to compensate for the modified input. Kind of a bad design, as the unit doesn't even fit in to the angled Strat style output jacks + the unit just falls out of my Rickenbacker if I try to actually use it in a live situation. Aside from Ron Neely's corrections, the schematic on this site is for this original plug in unit, which in it's original uncorrected form, was riddled with errors. The wiring will work either way. THe only difference is that with this schematic you will run the stereo jack as your output + your battery will switch off when you pull the output cable. With the Beavis Audio schematic, they were thinking in terms of a modern floor stompbox where your input is a stereo jack + the unit will power off when you pull the input cable.

bootle · Post by **bootle** » 26 Dec 2010, 23:52

acidfuzz wrote: ...However, my early builds from the schematic floating around were all a bit pickier than the OG VoxRPC.

...Essentially what's going on is that the schematic shown on this site is from the original Vox/Thomas Organ schematic for the plug in V809 unit...

...Aside from Ron Neely's corrections, the schematic on this site is for this original plug in unit, which in it's original uncorrected form, was riddled with errors.

hi, thanks for the info.
but which schematics are you referring to here?
the analogurru one posted in the first post?

ddpawel · Post by **ddpawel** » 31 Dec 2010, 10:49

I want to know something about R7. Will a different value for this resistor will change the shape of the signal? Can it act as a "spacing pot" in effect "Tremulus Lune"?

ddpawel · Post by **ddpawel** » 12 Jan 2011, 17:05

Here you can see the nature of the oscillator in this effect:
http://talkingelectronics.com/projects/ ... 22-41.html
Page 24

dmk2113 · Post by **dmk2113** » 24 Jan 2011, 18:04

For Q2, on Analogguru's schematic it says to go for a BC108A or similar transistor. Should I be aiming for one that's similar in terms of voltage, or similar in terms of gain? I have a veritable boatload of NPN transistors, but none of those, and I've only got one tranny socket which I'm reserving for the UJT

In terms of what I do have: 2N3904, 2N2222(+2N2222A), 2SC1000(Gr - crazy gain), 2SC1815, 2N5088/5089, BC 107-109 (all are in the B range), BC 337, MPSA13/18, and probably one or two that I've missed.

Obviously the closest thing I have to a BC108A would be the BC108B, but the ones I have all measure 300+ Hfe and if the gain is crucial here, then maybe I should go with something else? Or does it not really matter...

R.G. · Post by **R.G.** » 24 Jan 2011, 18:45

dmk2113 wrote:For Q2, on Analogguru's schematic it says to go for a BC108A or similar transistor. Should I be aiming for one that's similar in terms of voltage, or similar in terms of gain?

I'd say gain. I got one of these running over the holidays - well, OK, my special version with a lot more of the politeness we expect from pedals these days, like no switching noise, no volume loss, no ticking sound in the signal, 2N6027 PUT instead of UJT, etc. - and did some messing with modulator transistors. For the devices I tried, more gain was better, as a broad generality. I liked 2N5088, 2N5089, and 2SC1815.

Remember that what really matters is not the transistor "beta" gain, which is usually specified, but "alpha", the reverse-mode current gain. That's linked to the forward-mode current gain, but is not something the makers talk about much.

I'd say put in the 5088 or 1815 and only go to something else if you're not happy with the sound.

By the way, eliminating ticking requires thinking about where the current is going, not what the voltages are.

dmk2113 · Post by **dmk2113** » 03 Feb 2011, 09:40

Thanks RG - I don't know where we'd all be without you.

I'm happy to report that not only does the circuit work well, it worked well with all 11 of the UJTs -culled from 2 different sources - that I tried in it... all of them provided pretty much identical results. As for the NPNs, I tried BC108s 2N5088/5089, a bunch of Darlingtons, some SC1000s and a few others. They all actually sounded more or less the same, but the 2N5088 seemed to provide the best results.

I am curious about one thing, though - does anyone know of an effective mod to increase the depth of the effect? Before adding a depth knob I tried different valued resistors at the input stage before settling on 22k, which seemed to provide the right balance of depth and volume, but since adding the extra pot I've found that I wish it could be just a little bit choppier. I tried different value pots at that stage but none of them had too noticeable of an effect (at 2k there was a huge volume drop and at 100k about 7/8 of the pot's rotation simply produced a dry signal. Are there any other components I should be looking to modify?

Ordinarily I would just breadboard it and try a bunch of things, but I need to get a new breadboard

(I have zero background in engineering/science, and while I've figured out pretty well how my fuzz pedals work, modulation circuits are still mysterious to me)

Vox - Repeat Percussion (V 809) [schematic]