freestompboxes.org

analogguru - bump!

Ooops, I missed something.....
I will answer you tomorrow morning - too tired now.
Please send me the better quality schematics by mail.

thanks,
analogguru

bajaman wrote: Regarding brake function.
If I used a dpdt switch and connect it so that it breaks the wipers to R104 and R107 of the frequency trim pots would this have the desired effect of gradually slowing both oscillators to a standstill condition?

thanks
bajaman

Bajaman, with my old Leslie (three speed) I don't really remember a slowing down effect when off. On a three speed unit it's a very quick off (brake used), not like the effect of going from fast speed to slow. So having true bypass switching and having the effect at unity gain with the bypassed signal would achieve very much the same effect.

The three speed units were built more recently (non-tube). The really popular units like the 122 and 147 (tube) only had 2 speeds, fast (tremolo) and slow (chorale). Older models had 1 speed fast (tremolo) and a off that had a brake on it so it came to a halt quickly (like the three speed). These 1 and 2 speed units would be the ones you will most likely have heard on popular recordings over the years for both organ and guitar.

Here's a fantastic site for a lot of info about Leslie's mechanics, speakers, crossovers etc.

http://www.captain-foldback.com/Leslie.htm

Thanks Vanessa
Perhaps brake function is the wrong way to describe what I want to achieve, which is a gradual slow down of both rotors to a standstill condition - freewheeling until the spinning gradually stops.
i have a Leslie single speed cabinet in my workshop on semi permanent loan (makes a handy bench for dismantling amps on ) - I think it is a 122 model - I completely rebuilt the tube power amp in it - uses 4 x 6V6 tubes for 20 watts of pure tube distortion
I had some old nos black plate RCA tubes that I used , so you can sure guess how it sounds .
it has been very useful for AB testing against all the rotary sims - pity it is only single speed though.
Thanks for the link - I will have a look see
cheers
bajaman

Sounds like a cool amp. Black plate 6v6's... Tone monster for sure!

Those single speed Leslie's are pretty cool in and of themselves. The author of that site goes into some detail about them and how he found out about them. It pretty much fits the way I found out about them and I love them now too.

bajaman wrote: Perhaps brake function is the wrong way to describe what I want to achieve, which is a gradual slow down of both rotors to a standstill condition - freewheeling until the spinning gradually stops.

check out this link:
http://www.geofex.com/FX_images/ramp-lfo.pdf
the bottom schem.
regards
P.

So tech-talk will follow...
In the meantime thanks bajaman a more readable schematic of the leslie part:
Korg CX-3 Leslie schematic (KLM-246)

analogguru

pz wrote:

bajaman wrote: Perhaps brake function is the wrong way to describe what I want to achieve, which is a gradual slow down of both rotors to a standstill condition - freewheeling until the spinning gradually stops.

But Leslies's don't do that. They brake quickly on both the 1 and 3 speed versions that have an off ability.

Bajaman, how about the noise performance of this circuit? Is it noisy?
Replacing 3004 with 3007 and doubling the clock freq should reduce the clock feedthrough. But there are no typical BBD anti noise tricks like pre&deemphasis or companding. A 5-10k trimpots instead of R18+R19 and R60+R61 pairs could be added to clean up the signal a little bit.

How about replacing the present VCA with some OTA like 13700? The second ota could be used in the other channel.

Analogguru, thanks for the bigger schem.

Piotr

Bajaman, how about the noise performance of this circuit? Is it noisy?

It is a little bit on the noisy side - but certainly a lot quiter than a genuine Leslie speaker cabinet .

How about replacing the present VCA with some OTA like 13700? The second ota could be used in the other channel.

Why? - I do not see any problem with the VCA and the bass section does not require it as the sound tends to be omnidirectional in that section.
I found that only a very small amount of amplitude modulation was neccessary to achieve a realistic high frequency effect. Incidentally I lowered R29 (4M7) to 4k7 to get this section to work properly .

I suppose I could totally redesign this effect by including the pre emphasis, de emphasis and compander circuits used by Roland in their Boss DM2 or Ibanez AD90 units, but my initial idea was to build a clone of the original to evaluate it's performance with a guitar - thanks for the helpful comments regarding this design and it's shortcomings.

I am planning to redesign this in the future - I plan on using the Princeton Technologies PT2399 digital delay chips etc., but too busy with other projects at the present time
bajaman

bajaman wrote:

Bajaman, how about the noise performance of this circuit? Is it noisy?

It is a little bit on the noisy side - but certainly a lot quiter than a genuine Leslie speaker cabinet .

How about replacing the present VCA with some OTA like 13700? The second ota could be used in the other channel.

I think it's pretty perfect as is (the CX3 Leslie was not half bad). When I build it I might tack that little distortion circuit they have for some grind. Instead of stereo out, I think I'll split the mono out via a crossover @ 800Hz to get a quasi stereo effect high and low (like the original) unless there's a better way of doing it?

Hi Vanessa - it is definately worth while running it into a distortion unit (after it!!! )
cheers
bajaman

Why? - I do not see any problem with the VCA and the bass section does not require it as the sound tends to be omnidirectional in that section.
I found that only a very small amount of amplitude modulation was neccessary to achieve a realistic high frequency effect. Incidentally I lowered R29 (4M7) to 4k7 to get this section to work properly .

What i meant was to use the second ota in the stereo mode, not in the lowpass section. But my problem is, sometimes i like to make things too complicated

I suppose I could totally redesign this effect by including the pre emphasis, de emphasis and compander circuits used by Roland in their Boss DM2 or Ibanez AD90 units, but my initial idea was to build a clone of the original to evaluate it's performance with a guitar - thanks for the helpful comments regarding this design and it's shortcomings.

Ok! With al those add-ons i suppose it would grow up to a rack project

I am planning to redesign this in the future - I plan on using the Princeton Technologies PT2399 digital delay chips etc., but too busy with other projects at the present time

I see one problem here with 2399. Originally the circuit uses BBDs 3004. They give delay time range from ~2.5ms to 25ms according to datasheet. This range could be widened with some tricks, of course paying the cost of narrower bandwidth in case of longer delay times.
Taking a look at the 2399 datasheet shows that the minimum delay time is about 30ms and that is the upper limit of the mn3004. And that happens in 2399 with the R=0.5ohm (setting the clock frequency) the clock operates at 22MHz. I suppose even if we could push the VCO to go further it would saturate and the clock waveform would be far away from square.
So i think that 2399 isn't a good idea, it has a different delay range for this effect.

Thanks for the schematics, now it is easier for me to figure out the correct component numbers. Since I need (to avoid the Zachieeeee-Syndrom) a break from the flood of booteekers and their worshippers streaming in currently - here is the promised techtalk. First I have to say, that I am not an expierienced technician, so if you need any further assistance maybe Brian M(arshall) from Subdecay records could be the right person to talk to.

Okay - now i understand you analogguru - I duplicate the virtual earth mixing resistors R94, R95, and R97 and invert the control voltage from the hf VCO to modulate the duplicate VCA (Q5, Q6) feeding through the duplicate C28 (10uf). Sorry I misunderstood your original post.
Would an inverting op amp with 2 x 10k resistors be suitable for the CV inverter stage?

Should work, ensure V/2 at the non-inverting input.

Looking at the ramp up and down times, am I correct in assuming that the CR time constant for the bass section is 4.8 seconds?
The treble section time constant is 1 second?

I didn´t calculate or measure them but the values seem reasonable. Treble is the faster...

How does D3 in series with R47 affect this time constant? Does it make the speed change from fast to slow quicker than from slow to fast?

Yes one direction is faster, I assume it´s the speed up-time

My prototype seems to slow down too quickly - if i increased R47 or R46 would this give a longer time between slow and fast speed change?

I assume R46 for down and R47 for up....

Anyone got any ideas on how to slow the oscillators down to a stop condition (brake function)
Cheers
bajaman

Theoretically it should be possible to the stop condition, but....

Analogguru - your assistance again please.
Regarding ramp up and down times for the hf section.
Am I correct in assuming the the time taken to ramp up speed from slow to fast is the time constant of R46 and C17 - 1 second, and the ramp down time constant is the time constant for R46 in parallel with R47 and C17 - 0.24 seconds, or is it the reverse or something completely different?

See above.

Regarding brake function.
If I used a dpdt switch and connect it so that it breaks the wipers to R104 and R107 of the frequency trim pots would this have the desired effect of gradually slowing both oscillators to a standstill condition?

thanks
bajaman

You can replace Q12 by a stompswitch if you like.

The circuits around IC6/Q7 and IC6/Q11 are standard VCO-schermatics. They are used widely for example also in the Coron DS-8. If it should be possible to run them to a stop condition, then only with zero volt at R43/R45. Problem can be that oscillation disappears suddenly.

I am planning to redesign this in the future - I plan on using the Princeton Technologies PT2399 digital delay chips etc., but too busy with other projects at the present time
bajaman

Argghhhhh....

IMHo the use of MN3207 should be possible, you only need to introduce a stabilization fot the 4046 and the MN3207 (5V or 6V ?) and maybe a different bias.

Time for a break.....

Analogguru

If you want a unit with the impact of a standard electric mistress, forget companders and heavy filtering, and just worry about layout and decent clock buffering. If however you want a namby pamby Boss type chorus .................

The CX3 "leslie" is not a "namby pamby" bossalike, but actually a decent mono slow-speed leslie simulation. It sounds pretty decent on the organ. I don't know how it sounds with guitar. There are tweaks and ideas mentioned in this thread for improving the fast speed sound, making it stereo, etc.

Stephengiles wrote:If you want a unit with the impact of a standard electric mistress, forget companders and heavy filtering, and just worry about layout and decent clock buffering. If however you want a namby pamby Boss type chorus .................

Jeez, I need to pay better attention. I just now saw this thread. Super job Steve.

Thanks Clay - and all the others here that help and inspire me
bajaman

First post here, so I say hi to all of you.

Bajaman I think you did a fine jobe with the CX3 leslieboard.
I did about the same thing with this board, but I like yours better especially since you managed to make it smaller.

Since I'm not yet allowed to post URL's I can't show it right now.

I've just started building one using your layout, so I hope to report back soon with a build report and soundclips.

André

I just finished my CX-3 leslie and I like it very much.

Very usable effect for guitar.

I did 2 mods to the circuit.
I changed the 3.3k resistor in the ramp up/down circuit of the horn part
to 100k to increase the ramp down time.
Actually I think the schematic says 33k instead of 3.3k, but I prefer 100k.

I also added a RC lowpassfilter just before the CV input (pin 9) of the horn's
clock generator to decrease the LFO sweep in tremolo mode, because I think
it sounds more musical this way. I used a 1M trimmer and a 220n capacitor for this so I can dial in the right amount of sweep.
The filter has no noticable influence in chorus mode.

I will do some soundclips as soon as I boxed this thing.

André