Assistance needed modifying Tremolo circuit

[antonis]

it turns out that big old 470uf cap effectively completely disables the depth control.

Is that so..??

[ElectricDruid]

Incidentally,,, how does one go about obtaining the knowledge to determine what values = what modulation through this circuit?

In general terms, with stuff like this you're looking for an RC time constant - a capacitor charging up/discharging through a resistor. For LFOs, the cap is usually pretty easy to spot because it's nice and big - often 1uF or more. The resistor can be harder to see, but if there's a rate control, then that's usually it, and it often has another series resistance in circuit too.

Is there a formula for this, or can you model the circuit in Eagle or something (says he, pretending to know how to use eagle...)

There probably *is* a formula, but that's not how we mostly do it. There's a lot of other factors can influence the specific frequency a given RC combination will give in an LFO circuit (like schmitt trigger threshold voltages, for example) so rather than try and work out all the details, we can work from where we *are* and see where we're *trying to get to*.

So we look at the range we've got *now* with the known values. If you double the RC time constant, you'll halve the frequency. That can be done by doubling the resistor's value, or by doubling the cap's value. Doubling both would drop the frequency by a factor of four, two octaves. Likewise, halving the resistor or cap values will double the frequency. You can go up or down by a factor of two, three , four, whatever you need.

That gives us a couple of rules of thumb to use:

(a) Find the"R" and the "C" of the RC time constant
(b) Adjust the frequency by multiplying or dividing R or C by a chosen factor to get where you need to be. Remember to think about both the minimum and the maximum value when dealing with pots, or pot+series R.

Some stuff isn't this simple, but it's a good place to start. Clearly there are lots of ways this can *not* work, but I bet this is what most of us do about 80-90% of the time in situations like this.

[jhaneyzz]

Yarrrrgh...

While making the changes suggested by pacealot. i happened to notice that I had a 47uf cap in C6 rather than the 4.7uf specified.

So.... ordered a new c50k pot to replace the original which i broke.

I'll put everything back to original specs and see where I am before moving on.

oops...

[jhaneyzz]

wow, that is probably the best answer I've ever gotten to any question i've ever asked.

super helpful.

Incidentally,,, how does one go about obtaining the knowledge to determine what values = what modulation through this circuit?

In general terms, with stuff like this you're looking for an RC time constant - a capacitor charging up/discharging through a resistor. For LFOs, the cap is usually pretty easy to spot because it's nice and big - often 1uF or more. The resistor can be harder to see, but if there's a rate control, then that's usually it, and it often has another series resistance in circuit too.

Is there a formula for this, or can you model the circuit in Eagle or something (says he, pretending to know how to use eagle...)

There probably *is* a formula, but that's not how we mostly do it. There's a lot of other factors can influence the specific frequency a given RC combination will give in an LFO circuit (like schmitt trigger threshold voltages, for example) so rather than try and work out all the details, we can work from where we *are* and see where we're *trying to get to*.

So we look at the range we've got *now* with the known values. If you double the RC time constant, you'll halve the frequency. That can be done by doubling the resistor's value, or by doubling the cap's value. Doubling both would drop the frequency by a factor of four, two octaves. Likewise, halving the resistor or cap values will double the frequency. You can go up or down by a factor of two, three , four, whatever you need.

That gives us a couple of rules of thumb to use:

(a) Find the"R" and the "C" of the RC time constant
(b) Adjust the frequency by multiplying or dividing R or C by a chosen factor to get where you need to be. Remember to think about both the minimum and the maximum value when dealing with pots, or pot+series R.

Some stuff isn't this simple, but it's a good place to start. Clearly there are lots of ways this can *not* work, but I bet this is what most of us do about 80-90% of the time in situations like this.

[pacealot]

Yarrrrgh...

While making the changes suggested by pacealot. i happened to notice that I had a 47uf cap in C6 rather than the 4.7uf specified.

So.... ordered a new c50k pot to replace the original which i broke.

I'll put everything back to original specs and see where I am before moving on.

oops...

In fairness, it was d_a's idea before mine! Credit where it's due.

But now we know why it was going so slow. Good catch — I think we've all made that variety of mistake at some point. And now, if the stock settings still don't give you the LFO range you'd prefer (and I understand as I'm pretty persnickety about trem ranges myself), now you have some ideas about what to change to get you there...

[stonerbox]

Is there a formula for this, or can you model the circuit in Eagle or something (says he, pretending to know how to use eagle...)

I did it from my hazy memory from tinkering with old builds but you can also simulate the circuit in LTSpice.

To the left is the original and to the right is the altered version. The altered version produces a sort of square wave and can go much faster than the original.
The following changes to the pendulum tremolo schematic (not my image below) is:
Lower R13 to 22k.
Lower R14 to a 220Ω.
Swap the C50k pot to a 25k (if you feel the need to limit how slow it can go).

edit: sorry for the huge image.
[click for full size]

[duck_arse]

doesn't R15 provide the positive feedback the oscillator needs to run? will it work shorted in the real world?  and the 220R stop resistor against 25k plus 220R is going to make it harder to get the frequency you want compared to a 50k and 50k plus 2k2. in my opinion.

maybe a 125:1 range verses a 25:1 range, pardon my inaccuracy.

[stonerbox]

Listen to the man!

doesn't R15 provide the positive feedback the oscillator needs to run? will it work shorted in the real world?  and the 220R stop resistor against 25k plus 220R is going to make it harder to get the frequency you want compared to a 50k and 50k plus 2k2. in my opinion.

maybe a 125:1 range verses a 25:1 range, pardon my inaccuracy.

[antonis]

doesn't R15 provide the positive feedback the oscillator needs to run?

Actually, R13 does it..

C5/R15 brach is employed as the differentiator part of LFO..

[duck_arse]

C5/R15 brach is employed as the differentiator part of LFO..

which it needs to run [as intended] ? pardon my any inaccuracy.

[antonis]

 

Come on, Stephen.. 
(I was just pointing positive feedback resistor proper numbering..)