4 pt2399's in series / long clean delay..

[SISKO]

It's probably an impedance mismatch again. Got a recent schematic?

+1.

I dont have any basis to support this but here it goes what i think:
Opamp output impedance is low, C6 has low impedance too so the feedback path sees a small impedance to ground at the opamp output when the JFET is on, therefore you set your repeat knob accordingly.

When the the JFET is off, theres an infinite impedance to ground at C6, and this modifies the point where the repeat level was set because all the impedance mismatch.

IF this would be the problem id try:
A resistor to ground at the C6 -FET juction
Reduce C6 to 100n as in the ST. This would raise the impedance when the jfet is on and make the impedance mismatch less noticeable.
Replicating the feedback-mix path as in the ST.

[deadastronaut]

cheers sam an sisko...

i'll give that a go...thanks guys.

edit: tried that, but i get a massive bassy thump when switching...hmmmm..

[SISKO]

What did you tried? Those were singular options, not all at once. Sorry, i was not clear

If you choose to try the first one, the resistor should be small enough to make not noticeable the mismatch but big enough not to modify the DC level at the juction C6 - Jfet. If  the source and drain are not at the same DC level, it will POP

[deadastronaut]

hi,

the 100n on its own just changes the tone of repeats...

a 100n with resistor to ground makes a thump when switching

a resistor to ground from c6 + makes a loud thump when switching.

i tried 100r to ground, stil a thump, and lowers the repeats levels massively.....

[deadastronaut]

well i have the repeats on a much better equal basis now..thank @#$%..

had to adjust a few things, so the repeats didn't get thinner as it trails off...(which was a nice option too though..)

i now have good oscillation too, which doesn't get way out of control (louder) it retains its volume as it oscillates, really cool just using it as a noise machine with the modulation etc...anyway..

so the schemo posted will need some adjusting....again.

i nearly gave up on this more than a few times....my missus doesn't want to hear another word about 'delays'...and my cat hates me... 

but sod em, i'm soldiering on regardless..

[slacker]

Hi Rob sorry for not replying to your PM, for some reason the notification email ended up in my spam so I never noticed it.
I would have suggested the same as Sisko except I think the resistor should go to Vref rather than ground.
Another way would be to stick a buffer after the fet switch so that the change in impedance doesn't affect the following bits of the circuit.

[deadastronaut]

hi ian, no worries man, 

i went with how merlin did his st  repeats/mix controls section, just had to tweak it a little...

i'll get there.....eventually.

cheers man.

will have to re adjust the chip jumping values again, but thats easy enough ...he says.

[SISKO]

I would have suggested the same as Sisko except I think the resistor should go to Vref rather than ground.

my bad

Another way would be to stick a buffer after the fet switch so that the change in impedance doesn't affect the following bits of the circuit.

I tought of this too. Ive simulated the circuit and indeed shows two diferents singal levels depending on the switch state, but the level boost is not as big a i would imagine, a 10% maybe.
But im positive that this particular problem can be solved easily.

@deadastronaut you walk too long to quick now! I admire your persveration, cant wait to see this delya running

[deadastronaut]

cheers sisko,

yep I walk a very, very  loooooong road on every project i decide to '' embark'' upon ( sounds like an expedition, doesn't it,  which it is, for me anyway) ..... i'm certainly no EE, more of a compulsive  tinkerer really,


but i do have fun putting my own head through EXTREME TORTURE...to a final result...its great.  

...............

i'm glad you took the time, and had the curiosity,  to sim it man..welcome aboard the endless delay project!

Back soon...hopefully.

[deadastronaut]

my next little hurdle is getting the 'chip jumping' resistor values right.....not too low, and not too high..

when all 3 are 'jumped' there is an accumulative feedback difference, compared to straight delay, so its going to be a bit of a balancing

act to get them to be noticable and playable, without each 'jump setting' dominating the other....sounds cool though...almost there on that though.


my other issue..

currently i'm using a 1M pot for short delays...and then switching in a 1M resistor for longer stuff....but i'd prefer to have one switch less

i guess i would have to get  a 2M pot to get the full range of fast to uber slow...as switching in the 1M resistor will not let it go fast....

[deadastronaut]

ok, i'm nearly done with this...

i want to squeeze as much out of the lfo as possible though..

i want it fast AND REALLY slow ,(for messing with osc)  but without having to switch a bigger/smaller electrolytic to achieve it...hmmm big ask it seems eh...

hmmm...been tinkering, and simming with a crude circuit sim...

any ideas?..links?..to try..

edit:  nevermind, i'll crack on with what i have..

[Jdansti]

What range of capacitance do you need?

[deadastronaut]

hi john, heres the lfo i'm using at the mo,

i used a 1M Pot as i want a long sweep for when messing around with oscillation etc..

problem is the fast is good but is all bunched up the very top end of the pot....but it will sweep nice and slow too

the depth is fine, but that too is a bit bunched at the top end too...

any ideas?.. (i am using trimmers at the mo)

[samhay]

This is when using trimmers on the breadboard becomes an issue.

You almost certainly want to use a reverse log pot for the speed control (Tayda has them - http://www.taydaelectronics.com/potentiometer-variable-resistors/1m-ohm-c1m-1mc-anti-log-taper-potentiometer.html).
If that doesn't give you a smooth enough control, you could try to rig up a dual pot with the second wiper selecting between two caps (C51), but this is almost certainly more trouble than it is worth.

Likewise with depth - you may want to try a log pot.

[deadastronaut]

cheers sam, yeah i thought my 'linear' trims might be the culprit..

so reverse audio on 1M, and 47K log for depth...

ive read that i can use a 1M backwards?..will that work the same?

[slacker]

ive read that i can use a 1M backwards?..will that work the same?

An audio pot wired backwards gives the same sweep so you get the fine control of the fast speeds but the direction is reversed. So fastest is anticlockwise.

[deadastronaut]

cheers guys, i'll dig a log 1M out and try it...


edit: brilliant they work a treat, backwards , but thats fine...cheers guys..

[deadastronaut]

yep they both work great backwards...much better transistion/action...sounds great!..


ok, my last little issue..honest!


i'd like to make a 2M pot... instead of having a 1M pot and then switching in a 1M resistor.


as i have it, the 1M POT will give me fast to slow delay.....but then i add a 1M resistor in series with it to get longer delays, but will not go to the fastest with that 1M switched in...get me.

any little tricks to achieve this?...to save having a switch etc..?..

[samhay]

Would not the simplest case be to use a 2M log pot? Maplin has a plastic shaft one (FW29G).
More complicated options would be to use a dual-gang 1M (anti)log pot and wire the two halves in series, or to use a switched 1M pot where the switch e.g. shorts out your 1M resistor.

[deadastronaut]

yeah i thought a 2M would be the easiest... ...i'm anti plastic shaft though.  

the switched pot sounds like it might be useful for that and also the lfo too...good idea!...save on 2 switches.  


edit: just found some 2M pots...

and a 100K for my LFO switched, (should be ( 47-50k ) but its linear...hmmmm...Bummer.