Recreating the Paia 4720 VCO

First Posted July 2, 2024

4720 KiCad 8.0 Project File

I tend to make a lot of fun of Paia kits, and in particular the 4700 synthesizer kits.  However, for what they were, they are something to be admired.  The primary goal John Simonton had in mind when he designed these kits was to give the hobbiest something that did the job and was affordable.  And that he did.  If you want to see bare bones design, take a look at his.  You can download user manuals from the Paia website.


Sometimes John did make a boo-boo every now and then.  Sometimes it wasn't really a boo-boo, but the difference between using a $10 part and a $0.10 part.


The 4720 is a good example.  It is a ramp vco core.  It is very similar to the popular Terry Michals ramp VCO.  Rather than using a JFET to discharge the timing capacitor, the 4720 just uses a BJT (bipolar junction transistor).  And rather using an LM311 comparator to turn on the discharge transistor, a Unijuction Transistor is used to do that job.  Unijunction transistor are difficult to locate these days, but in 1975 they were very common.  Every thing else in the 4720, the ramp to triangle, pulse width modulator, the triangle converter are all not a resistor too many implementations.

It is hard to say , for me at the moment because I am too lazy  to check things out, if any of the cost cutting compromises were counter productive.  One that I really question is the use of the unijunction transistor.  I don't know how much the treshold level drifts with temperature, but that is a rather obvious potential performance issue.

If you have downloaded and looked at t he schematic of my reimplimentation of the 4720, I did not try to make things bare bones at all.  Again, that is because I am Lazy.  And a lot of things are very different.

First off, I use a Triagle VCO core. This is an OCD issue.  It makes a nice clean triangle which makes it easier to make a nice clean sinewave.  In reality, no body will notice the difference, but hey, it is cleaner.  The ramp is made from the triagnle using a synchronous detector.  What that really means is a circuit that can switch between positive and negative gain.  Yep, the ramp will have a litle blip in it.  You just can't get away from that.

You will also note that there are a few more pots on the front panel. 






I don't know about you, but I hate wiring up panels, and even more so these days with my extremely limitted eyesight.  So besides the main board there are also two auxilliary boards that have the pots mounted on them and another board that has the Banana Jacks.  Both of these boards have connectors that accept IDC connectors making it so it is posible to connect the boards together with nothing more complicated than ribbon cable.