Trying to understand ARP 2500 1004 VCO

[Ricko]

I haven't found a detailed explanation of ther ARP 2500 VCOs. Here is my stab, based on the schematics on guitarfool website. Is this about right?

We can think of the 1004P VCO as having four unequal stages: the CV input mixer, the expo converter and triangle core, the waveshapers, and the output mixer (not shown on the 1004 schematic.) The schematic is not really laid out like this (in the bottom right quadrant), but I think my approach is less confusing.

The CV input mixer is conventional. The output mixer can be conventional: it lets you mix the 5 waveshapes. We can also dispose of two other things that complicate the schematic: the Enable switch S1 that discharges the integrator caps and lets you turn the Vco off, and the Range switch S2 to switch in extra integrator caps for low frequency operation.

Next to the waveshapers: the triangle core supplies tri and square waves. There is a sine shaper and a comparator for variable pulse waveshapes. (No drama, but taking the pulse from the tri means PWM will have a vibrato or chorus effect since the leading edge wobbles.) But now something unusual: the sawtooth wave is not generated by fiddling with the tri and square waves. Instead, the square wave through c15 resets a dedicated integrator A4 which is driven from a second tap pin2 of the 4001 exponential module. So we get a perfect saw.

Which brings us to the tri core. This has two exponential generators: one positive the other negative. The expo generators can each be enabled or disabled (pin 1 of the 4001 and pin 2 of the 4002.). The tri core has a conventional integrator A3 and comparator (see around Q12). The squarewave produced by the comparator toggles the positive and negative exponential converters using the enable pins. Thus the up/down cycle of the triangle.

anyway, that is how I read it today...
Cheers
Rick

[Ricko]

Aha. So now I think I can also figure out the other impenetrable VCO schematic that has had me long stumped: the Aries 317 (schematic can be found at the modularsynthesis site.) probably obvious to everyone else...

It is pretty much the same as the ARP 1004: a tri core vco with a sawtooth generated by an integrator using a second tap on the expo converter. (The comparator for the tri core is U4 with Q5,Q6. The reset for the saw integrator is through c9.)

The main difference is that there is only one expontial converter, and a CA3080 U2 is used to toggle between positive and negative to feed to the tri cap C16

(What does not make sense in the schematic to me, is why the control pin 5 of 3080 U4 is shown put to ground. I thought it would need to be some + or - value? Is that a mistake perhaps? One leg of a voltage divider omitted perhaps? Or is it something for 3080 used as a comparator? )

Update: The Aries LFO is another tri core with this same method of generating the sawtooth.

Update 2: Now I found a Theory of Operation pdf at CEM 3374 which confirms my take on the tri core. Nice. On the issue of the pin 5 I note that the resistor is the only one not labelled on the schematic: presumably r22. I looked at the 3080 app note, and when It use 3080 as a comparator, they supply a current to pin 5, at least in the examples, so I still dont get it.

[piedwagtail]

I note Dennis Colin designed a lot of the Arp2500 and then left to be the lead designer for Aries, so this double use of the exponential converter is probably his signature.

Regarding the CA3080, it's still getting current from ground because pin 5 sits at -14.4V according to Thomas Henry's 3080 Cookbook.( Jung has the pin up to + supply in the IC op-amp cookbook comparator). As part of the lore of 3080s one needs to consider the current sinking into virtually the negative supply voltage for Iabc calculations.



Robert

[Ricko]

Ah thank you!

I have seen the multi tap expo converter on vcfs before, but hadnt considered it would be useful for vcos.

rick