Rather tacky that Rob Hordijk’s name is mentioned nowhere in their description.
2 Likes
I sent the links to Biyi.
Biyi is the heir of Rob’s Legacy. He has started to continue the brand as Biyiblip and recently released the Gibber v3 (basically a rungling speech synth)
Have you guys seen or played with the Peterlin or Golden Section???
These were okay’d by Biyi 
Golden Section is like the Peterlin on steroids lol - has x2 sample and holds and x2 theremin style CV controllers…
I just recently got my Golden Section in and its about as much fun as my big dual benjo lol.
Figured I would mention these since a different kind was mentioned - but this one was approved
(Ben2 was also approved before Rob passed)
1 Like
My microscope practice was 0201 at 5GHz. But I tend to balk at anything that starts off with “04”.
I totally agree its a schmozzle…
I also weirdly like it. To me it mimics the chaotic heart of the Benjo
analog output kosmo version would fix that 
At 20 x 40 cm, probably.
it certainly would be a lot more finger friendly and hopefully easier to build, the ben2 looks a nightmare to hand solder.
I will be hand soldering so yeah … FUN!
I’ve been breadboarding and simulating the rungler section of the Benjolin and…
Here’s the schematic:
The left op amp output is labelled XOR. So I expected it to be an exclusive OR of its inputs, but that’s not what I’m getting, either on the breadboard or in the simulation. I’m getting the output to be high or low when the P1 input is low (-9 V) or high (+9 V) and Q8 is high (+9 V), but high regardless of P1 when Q8 is low (-9 V):
| P1 | Q8 | Output |
|---|---|---|
| 0 | 0 | 1 |
| 1 | 0 | 1 |
| 0 | 1 | 1 |
| 1 | 1 | 0 |
Is this correct? It sort of makes sense to me, because when Q8 is low the transistor is reverse biased and I wouldn’t expect much of anything to happen in response to what P1 is doing. But this isn’t an XOR truth table. In fact it’s OR (NOT(P1), NOT(Q8)).
are your boards or gerbers available somewhere for this? Looks like a fun project.
Anyone (@Rumblepants, @d42kn355, @lookmumnocomputer, anyone else) who’s built the Benjolin or a standalone Rungler: Do you find the loop offset pot (as seen in the Kweiwen, Forestcaver, and BE2N versions) useful?
In my breadboarding what I find is that in the middle of the pot range it gives the original behavior; at either end it replaces the input signal with either ±9 V: +9 V results in looping the shift register with no modifications, so the output CV loops, while -9 V results in writing all ones into the shift register so the output CV just goes flat. (If the XOR signal really were an XOR I think the -9 V would cause double looping of the shift register, inverting and re-inverting each time.) But in between it doesn’t seem like it does — and from the schematic it doesn’t seem like it should do — anything in between those behaviors. Not like the Turing Machine where it goes from double looping to random to looping in a somewhat gradual way.
So in fact it seems like it’d make more sense just to have a toggle switch that connects or disconnects +9 V, giving either looping or standard rungling.
Or am I missing something, or seeing wrong behavior?
oh gosh. i cannot for the life of me remember. i cant seem to find my benjolin panel. i traded one for something a while back and not sure where the other one got to (it was in a case then out of a case and so on!) but the first benjolin is at the museum. i will double check
but off top of my head i remember there being more things happening inbetween. but i may be wrong. armed with your comment ill go twist it and see. watch this space.
2 Likes
Found my XOR problem! I’d looked at the schematic and noticed the transistor is unprotected against reverse biasing the base-emitter. So I did what I’d normally do (in the simulation and on the breadboard), I added a protection diode from base to ground. Tonight I said “that doesn’t cause trouble does it?” and took it out of the simulation and boom — now the output goes low if both inputs are low. It’s a real XOR.
But it bothers me! It’s deliberately sending -9 V (well, okay, about -7.5 V taking the TL074 output swing into account) to the base and relying on the transistor to do… whatever the sim is saying it’s doing. I’m no transistor expert but back biasing the base isn’t something one normally does, right? And the datasheet says the minimum emitter-base breakdown voltage is 6 V. That’s minimum, it’s probably higher, but relying on that is… a thing I don’t like.
On the breadboard it seems to work. And it affects the loop offset knob behavior: Now instead of shutting down the output when counterclockwise, it loops the shift register. (Double looping occurs at the clockwise end.) There also seems to be a spot near full counterclockwise that reliably does shut down the output; other than that I still don’t see much evidence of different behavior at intermediate positions.
Added next morning: On further investigation it seems the behavior when you vary the loop offset is rather more complicated than I thought. In particular, depending on the loop offset, the “high” level you get when P1 is high and Q8 is low can be different than the “high” level you get when P1 is low and Q8 is high, and similarly with the two “low” levels. Then if for instance one “low” level (but not the other) becomes higher than the shift register threshold it’ll change the behavior. What I see on my breadboard is: Near counterclockwise you get 8-step looping; somewhat above that you get nothing; above that is standard complex behavior; closer to clockwise the CV output pattern becomes simpler and more repetitive, but it does change sometimes; fully clockwise you get 16-step looping. So there are five ranges, four of them useful.
1 Like
4 Likes
This has me excited, wasn’t able to get a Kosrollz and have been watching everything that @d42kn355 has been doing now that he’s fully dove into Ciat-Lonbarde.
Best of luck with SLM if you read this CrucFX. Would love to come up to Appleton if you ever do any in-person performances or demos.
1 Like