MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Arp JFET binning / Note 11?
MUFF WIGGLER Forum Index -> Music Tech DIY  
Author Arp JFET binning / Note 11?
devinw1
Hey all... another Odyssey related question.. I'm looking at the B1 (oscillator board) assembly drawing and there is a "Note 11" for Q6 and Q13 for VCO1 and 2, respectively. Does anyone know what this note is? Or have any idea what characteristic Arp would have been selecting these FETs for?

The BOM lists 2 seperate PNs for the same tranny (2N5459) so presumably these are the binned ones:

fuzzbass
There are details missing here such as which version of the Odyssey you have. The VCOs in the TTSH share circuitry with many, but not all, Odysseys. The 2N5459 JFETs in the many TTSH I built were not selected. Any unit from the bag works, unless it's bad one or gets cooked going in (which happens more with FETs than say BJTs). My 2N5459s were all sourced aftermarket (usually UTSource).

That said, there may have been good reason for selecting FETs back in 1972. Manufacturing tolerances were much looser then. But, I would not worry about this note unless you found some new old OLD stock JFETs for that vintage sound.
guest
on the schematic, Q3 is labelled as 2N4392, so it might be a typo.
devinw1
Oh, BTW this is for the MK1 (B1 board).

Also, FWIW, I have replaced a couple of those 5459s on a different oddy where they were blown and just grab bagged it and it worked fine. Just curuious if they were binning in relation to my new project (A fully cloned replacement B1 board for my MK1).



guest
ok, i see now. and Q3 has a "see note 10" next to it as well. there was probably some internal binning, as those transistors are used differently, and the spread on the 2N5459 is pretty large. Q6/13 are being used as switches, so low off current and fast switch time would be ideal. this could be checked via the reset pulse on the VCO. for Q3 its a follower, so low gate leakage and and Idss on the lower side is probably desirable. that transistor could end up clipping the sampled voltage if Vgs got too large.
guest
if you have an original, you can measure the Idss and see what they are binned like (sample size of 1 aint great, but better than nothing).
devinw1
guest wrote:
ok, i see now. and Q3 has a "see note 10" next to it as well. there was probably some internal binning, as those transistors are used differently, and the spread on the 2N5459 is pretty large. Q6/13 are being used as switches, so low off current and fast switch time would be ideal. this could be checked via the reset pulse on the VCO. for Q3 its a follower, so low gate leakage and and Idss on the lower side is probably desirable. that transistor could end up clipping the sampled voltage if Vgs got too large.


Cool, appreciate the response. I'll have to see if I have an original. I may not have kept it, and the only B1 board I have now has been reworked by someone else to oblivion (hence me re-popping a new one...). But I know that Q3 on VCO1 currently works fine as far as the oscillator works well and tracks etc... And the sample and hold works just fine so I have that one as well.

Looks like Q3 is using the same binning criteria as the 2 high impedance buffers on the 4023 filter module, based on the Arp PN (A2803-003-1):

http://www.till.com/arptech/pdf/4035.pdf

And Q6/13 are A2803-003-2.
devinw1
So, I finally got around to measuring the 5459s. First off, I didn't find anything very different about the one from the poorer tracking VCO than the good tracking VCO, but this board has other issues so I'm just going to move on to my brand new board and put the ones I've measured on it and see how it goes. Here are the measurements. I measured Vgs(off) with both 9V (battery) and a 15V power supply and got roughly the same numbers:

VCO1 (tracks well)
Idss = 6mA
Vgs(off) = -2.35V

VCO2 (does not track well)
Idss = 5mA
Vgs(off) = -2.1V

I also measured the res of my stash of 5459s and got the following distribution of Idss (mA):
4
5,5,5,5,5,5
6,6,6,6,6,6,6
7,7,7,7,7
9

So, a nice little gaussian distribution centered around 6mA roughly. Also, not surprising that I had grab-bagged those other 2 and got a 5 and a 6, due to probability smile. For reference, here are the specs for the Fairchild version of this JFET:

Idss: Min 4mA, Typ 9mA, Max 16mA
Vgs(off) Measured at 15V: Min -2.0V, Max -8.0V

So, we are on the low end of both those specs for this batch.

I will report back on the new board VCOs in another thread when I finish stuffing it. Rockin' Banana!
guest
good work! which FET was that (Q6,3,13?)?
devinw1
guest wrote:
good work! which FET was that (Q6,3,13?)?


Yeah, so the one from VCO1 would be Q6 and the VCO2 one was Q13. Both are actually replacements, not the original. Unfortunately the VCO2 original I tossed, but the original (or at least old ass) one from VCO1 was dead when I got the synth so nothing useful to ascertain there.
MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 1
Powered by phpBB © phpBB Group