# Ring in the new

Bartender! A round of drinks on the house, if you please!

Yes, yes, I know I’m the only one here. Start pouring.

I have, ladies and gentlemen — okay, bartender and dog — this very day succeeded in building a Eurorack module with surface mount parts, unassisted!

It says RING on it, and it’s a dual ring modulator from PM Foundations.

A ring modulator is similar to a voltage controlled amplifier in that the output consists of one input multiplied by the other. The difference lies in the fact that in a VCA the second input typically is a positive control voltage, and a negative input is ignored, while in a ring modulator, typically both inputs are audio signals that can go positive and negative, and the multiplication is done for both positive and negative inputs. If both inputs are sin waves with frequencies $f_s$ and $f_c$, then the output is a sum of two sin waves with frequencies equal to the sum and difference: $f_s+f_c$ and $f_s-f_c$. Unlike with a VCA, neither input’s frequency appears in the output. Where things get really interesting is when one or both inputs have higher harmonics; then the output signal contains none of the inputs’ harmonics (unless the inputs are harmonically related), but does contain sums and differences of all the first signals harmonics with all the second’s. The result is a complex, non-harmonic tone whose sound depends on the relationship between the inputs’ frequencies. Put it into a VCA with an envelope with sharp attack and long release and you can get distinctly bell-like tones. I haven’t played with a hardware ring modulator since my undergraduate days with the ARP 2600*, and I’ve missed it.

This module has 1206 surface mount resistors and capacitors, two SOICs (a TL072 dual op amp and an LM13700 dual operational transconductance amplifier), a couple of surface mount electrolytic capacitors, two trimmer pots, a power header, and six jacks.

Instructions were pretty clear. Once again the new tweezers made assembly a lot easier… and once again my brain doesn’t work, because I put the TL072 in backwards. I caught the mistake immediately, cut the IC out and cleaned up the pads, and put a new IC in; at that stage, before mounting the taller parts, it was a lot easier than the IC replacement for the sample and hold. But I was understandably worried about whether it would work.

Once I got all the SMD parts in I had to take a break, because I’d ordered the trimmers and jacks from Tayda and they hadn’t arrived yet. I’d had a long wait. Tayda has 15% off sales frequently, and by frequently I mean in 2019 they announced one on average about every two weeks. The sales tend to last four or five days. So in late January when I’d put together a list of the parts I needed, I figured the next Tayda sale would be in early February, so I waited for that… and waited… and waited…

Of course that time of year is Chinese New Year, and Tayda is headquartered in Thailand and gets a lot of stock from China. And then there’s COVID-19, which has hit China hard, of course, and also Thailand. I don’t know if those are what kept Tayda from wanting to deal with the surge in business a sale would produce, but for whatever reason it was more than 30 days after the January sale announcement before they announced another one. And of course during those weeks my Tayda wish list kept growing… By the time there was a sale I had over 600 parts in my order. Mostly resistors, with a minimum order of 10 (at 1.2 cents each). Everything finally arrived yesterday.

I finished the module up, plugged it in, and tried it out. Good news: One modulator worked perfectly. Bad news: The other didn’t work at all. Or seemed not to, and then I got it to make intermittent noise. I convinced myself one of the input jacks was bad, and nearly performed a destructive removal on it before thinking to check it directly. Nope, it was making reliable connections to the cable and the rest of the circuit board. I spent a long time trying to track things down, checking connections on one modulator and seeing if they were the same on the other. (PMFoundations unfortunately doesn’t supply schematics, so debugging their circuits is a bit harder.) Finally I warmed up the connections on the second side of the TL072 and I think that’s what did it: the second modulator started working. I calibrated both, installed it in the rack, and celebrated.

* Off the subject, New York State’s passenger vehicle license plates are of the form LLL-NNNN, three letters and four numbers, issued sequentially; the first plates of this type, starting with the letter A (ACA-1000), were issued in 2001 and they’re now up to JLL-NNNN. The other day I saw a car with plate ARP-2857, or something like that, and it made me realize that surely there are or once were a couple of cars driving around with ARP-2500 and ARP-2600 plates, and I wish I’d had one of those.