Here’s mine:

7 Likes

First post from myself…
I have been pondering the rollz circuits for a few days after speaking to Rodrigo.

Has anyone thought of making an anti-roll? This could be achieved by swapping out the npn transistors with pnp’s.

As time = capacitance x resistance has anyone put a resistor in the circuit anywhere?

2 Likes

That’s an interesting idea, would anything else need to be altered or adapted on these circuits or just change the transistor type?

Yes, I’ve certainly noticed changes in activity rate if I clip resistors or capacitors etc across the Rolls nodes.

A pnp works the oposite way from a pnp. So I would imagine not. Definitely worth a go. It would only take 15 mins you knock up a four node roll. I don’t have a working rollz otherwise i would go for it.

As for a resistor you could just put it in series on the supply to each Roll. Then you could fine tune the speed as required. Or possibly have two busses for the supply for each of the resistor types.

2 Likes

I have some spare Rolls boards, I’ll give this a whirl and report back… Interesting ideas.

Oh and I have designed some dogvoice pcb’s in Eagle

8 Likes

That should be the 22K resistor, of course. Sorry for the typo!

I didn’t use an LED driver circuit of any kind, it was more of a trial and error method :wink:. I poked my multimeter around on the PCB, found the positive pulse behind the 22Ks and connected the LED on one of them via a 270R resistor. I used the same positive pulse for the orange outputs (and didn’t connect the regular output for that node). Since I am not an electrical engineer, it’s not unlikely that this is a very wrong way to do this, so don’t take my advice on this!

1 Like

When I tried the method that @sbaio describes, I found that the pulse which the led was connected to was shortened. I think this was due to the capacitor being discharged through the led.
That’s why I added the LED driver, which is the same as shown in @corpusjonsey schematic.

2 Likes

You could use a hex inverter if you wanted an output that was buffered. Or is the whole idea to chain the positive nodes together too?

Music from outer space has good examples of hex invert buffers and LED drivers.

2 Likes

I’m not sure if this was directed at me, but for me the idea behind the positive outputs was to have something akin to the orange outputs on the rolls of the Plumbutter: a positive voltage that’s related to the negative pulses. I never tried chaining the positive outputs of the rolls.

Thanks for the hint at Music From Outer Space. Never heard of the site and I love it. What a great resource!

Looking at the rolls circuits again, this makes a lot of sense. I think I need to upgrade my LEDs with a decent driver circuit.

@sbaio yes I guess it was directed at you. Was thinking of implementing a similar positive output. The only problem is the pulse would be a binary output. Which isn’t a problem for gate/trigger signals.

Ray Wilson from Music From Outerspace was a lovely guy RIP. I use his website a fair bit for inspiration and education.

I put together a 6-Roll board this morning using BC557C PNP’s instead of NPN’s and got no output on any of the nodes… Seems like something else maybe needs adapting, or I’ve made an error somewhere. I’ll do some proper checks with a multimeter when I get more time but as it stands… Nothing happening.

On a different Roll board I put a 50k pot and a 10k resistor in instead of a 22k and that does have some effect on the speed of the pulses. The speed change isn’t huge so I might mess around with different value combinations but certainly seems to work as you described.

If you look in the Roolz-Gewei / Labrolz paper:

it shows the circuits for the rolls that are in the Plumbutter. These are tempo adjustable rolls, but are more complicated circuits. I’ve modelled them in LT Spice, with a voltage divider emulating the potentiometer, and they work in Spice. I haven’t built them, though!

There are two parts to the circuit: one that contains the pot, two resistors, two PNP transistors and one NPN. This produces / restricts the current going into the rolls.
Then each roll contains a PNP and NPN transistor, two resistors and one (electrolytic) capacitor.
So, they are more complex circuits than the simple rolls, but should be buildable.

9 Likes

Just a thought - if I’m using BC557C PNP’s with electrolytics as hairy caps would they need to be reversed in order to accommodate the different transistors?

(I’m largely ignorant of how this circuit actually works, in case that isn’t obvious by now) :slight_smile:

@pmj sorry to waste your time. I’ll give it some thought over the next couple of weeks and think of a solution.

Glad the pot does something though. It’s a good start, for a simple hack.

1 Like

Anti Rando & Rando via 5-Roll cross patched…

(Includes a Rando literally retrieved from the bin thanks to the sharp-eyed @mlogger spotting I had the wrong chip installed…) :sweat_smile:

8 Likes

Has anyone built the Solardyne?

solardyne

It should be a piece of cake (it’s for Eighth Graders, apparently…) but not having any luck getting any sound out of it. Assume it’s a noise generator of some sort?

If anyone has got it doing something, what is the unmarked transistor? Seems like it should be an NPN or a FET?

1 Like

Has anyone build the MengQi PCB of Namastitar? I’m struggling to get my head around what to solder to the 10k ladder/swiper pins.

I haven’t made one myself but you could look at the paper circuit design. a Piezo? Or do you mean the wiper pot for the X resistor? If it’s the X resistor, I would try different ones by hand. Usually I start with 10k or 22k and work my way up to 100k,