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

Polyfusion Envelope Follower trouble
MUFF WIGGLER Forum Index -> Music Tech DIY  
Author Polyfusion Envelope Follower trouble
Jarno
It's a simple enough circuit, but I cannot seem to get the comparator to work. The gate output is always on and the LED lights all the time, regardless of fiddling with potmeters.
This is the original circuit:


And this is mine:



Layout, if that helps anyone:


I have added 100k from the base of the 3904 to ground, think that needs to be there. Also, I removed the 10M feedback resistor from the comparator. But both changes did not make a difference.

The envelope follower does work, but the comparator does not. Can't see anything out of the ordinary, other the comparator being the inverted kind (because the signal of the envelope follower is used inverted for the comparator).
Any ideas?

I did use LM324 rather than TL074.
Starspawn
Looks like the response switch should be an on-on-on type to allow for 3 choices. Is yours that as well?
KSS
Change R17 to 470ohm. Right now the comparator is against 470/471*15V=.03V

Put the 10M resistor back in. Its purpose is to create hysteresis so the comparator changes cleanly wilth slow input transitions near the trip point.

edit: remove your added 100K to GND at the Transistor base. Equation above is for the original 15V. Typo in the original circuit created your problem. They were going for a 5V trip point reference. Somebody put a K where an ohm should've been.
Jarno
Thanks guys, I have the reponse hardwired to a single setting, no switch yet.
I noticed the 470k resistor as well, I exchanged it to 22k, to arrive at roughly -0.5V, but I'll try 470r or scale the divider up a little bit (feels like a waste to burn 10mA of current for a voltage divider).
I'll also try the 10M resistor again, it was in there initially.

Wrt the 100k resistor from base to ground, there should be a path to ground somewhere, right?
KSS
Jarno wrote:
Wrt the 100k resistor from base to ground, there should be a path to ground somewhere, right?

There is. Right through the B-E PN junction.
KSS
When copying and modifying a circuit, it seems the best success comes in keeping it as original and get that version running. Then make mods to voltage rails, etc. That way you have a functional baseline to work from.

If you'd had a real PF module to work with, theis first step could be bypassed since you can probe and measure the workng module. Working from schematics, especially vintage schematics, it's nearly always good to do the circuit as-is first.
Reese P. Dubin
Following this because I have never gotten this one to work
Grumble
I don’t see the gnd connected anywhere in your pcb layout, or am I missing something?
tardishead
That's weird I bread boarded this a long time ago and it still works great
Straight from schematic
Maybe there was a different schematic at one point
Either that or I cant read properly and strangely put the correct resistor in by mistake
Ha!
Jarno
Thanks everyone, will need to do some more prodding tonight.
I did start from the un altered circuit (the schematic I posted is the circuit on the board and that's what I started from), it has top and bottom ground planes, and those are not filled in because of clarity, all of those grounds are connected in fact (Eagle made sure of it smile ).
Will need to record some voltages on a few points, maybe the comparator is working fine, the problem might be upstream.
KSS
Did you mean downstream? Upstream you said the envelope follower portion is working. Your added base resistor to gnd on the LED driver transistor will lock the comparator state. The trip point reference being at one end of the scale will lock the comparator.

The only thing between the working EF part and the comparator is the gate threshold pot. do you see the changing voltage of the EF output at the wiper of this pot? If yes, then upstream is not a problem.

Check again in the remaining portion that you have no solder bridges, missing net nodes or solder points. Incorrect parts? Have someone else read you the parts from the board, resistor colors. Do a greenline of the schematic from their readings It's easy not to see your own mistakes through over familiarity and 'I already checked that' bias.

Lift one leg of R20 and R21 to isolate the comparator output from its buffer via R20, and LED driver via R21. With all downstream parts offline, the comparator should work. If not, adjust the trip point reference resistors until it does.

Then replace the end of R21 (having removed your additional resistor). If the LED is not lighting with the comparator changes, then suspect the transistor, LED or R22. Once this is working move on to the output buffer.

Reinstall the leg of R20. Your change of the output resistor should not be a problem, but as mentioned before, it's always easier to build the circuit as designed and then mod. Not both at the same time.

Did you understand the answer about the GND being there on the other side of the PN junction?
MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 1
Powered by phpBB © phpBB Group