Slowing a Gobo right down

[LightKnitter]

Many thanks for the existing forum support here, it's helped me with sequencing colour fades for an art project that is working nicely.

I'm running unbranded RGBW twinkle LED engines via FreeStyler on my laptop. The twinkle is caused by a gobo rotating on the 5th channel.

The problem is that the gobo spins too fast even on the slowest setting, making the lamps look 'disco' rather than a very subtle, creeping shimmer I'm after. The gobo channel appears to have 4 settings: off, slow, medium and fast. I've tried CUE setups with 100ms increments of on/off on slow speed, which does work better than 'slow' but makes the gobo rotation jerky and maybe uneven. I've tried various 500ms fades on the gobo channel to attempt to reduce the 100ms minimum step but haven't managed a better effect than the on/off technique. (I don't know, but maybe I'm hitting the limits of the stepper motor increments even at 100ms on/off)

Anyway, just on the off-chance, is there some FreeStyler method I'm missing?

If not, any thoughts about swapping the stepper motor (28BYJ-48) with a slower one? (I'd be starting motor research from scratch.)

[buttza]

Hi,

Did you use the fader for that channel that is specific to the gobo?

The fader will give you more control. There are also other factors.
1. Stepper motor. (Can't go that slow)
2. 8 Bit / 16 Bit gobo speed adjustments.(this might be fixture related.)
3. Data transfer rates

More high end fixtures can be more precise with it's control that others.

[LightKnitter]

Hi Buttza, Thanks for those points.

Yup, the 5th channel with fade experiments (from memory) like 0 to 57, 25 to 60, 40 to 75 (The slow end of the 0 - 255 values). I don't know if I was missing the fade 'window' due to the 500ms minimum fade time against the 100ms minimum scene time.

I forgot to mention I tried speeding up the 100ms on/off steps by gradually increasing the sequence and global speeds. It got the the stage where the gobo stopped jerking but just became 'slow', as in the first rotate setting out of slow, med, fast.

I guess I'm going to look for a different line of RGBW twinkle engines or delve into stepper motor research.

[remco_k]

Swapping the stepper motor might not solve this for you.
It is the electronics controlling the lowest possible speed (=time) between every step. So swapping the motor for something else, would not change the stepping speed.

On the other hand, swapping the motor with a smaller step angle might render a solution. Although at that point the gobo isn't able to do a full rotation anymore, because the electronics aren't aware of this step angle decreasement.
Your current motor has 64 steps for 360 degrees rotation. That is 5,6 degrees rotation per step.
The electronics 'pulses' 64 steps for a complete 360 degrees rotation.
If you swap the motor with 200 steps for 360 degrees, then you have a 1,8 degree rotation per step. But, as I said, the electronics will send 64 steps for a complete rotation, but your motor then rotates only 115,2 degrees.
If its a real gobo (meaning it has to stop at an exact position), that might be a problem. If its only for a permantly rotating effect, then this might work. Just make sure to get a motor wit the same electrical specs.

Or, if there is enough space available, you can try and slow the rotation down with a tiny diy gearbox.

-------
Edit: I now see that your current motor already has a gear ratio of 64. So that renders a step angle of (5,625 / 64) = 0,0879 degrees. So it changes all numbers above. Swapping the motor for a 200 step version (and also have a gear ratio high enough) might still solve things for you.

[LightKnitter]

Hi Remco_k, thanks very much for all that info.

It's not a real gobo with varied segments around a wheel. As is, it just continuously turns. So far I haven't readily been able to find a similar sized motor to drop-in.

That's an interesting idea to rig gears outside the motor. I'll give that further thought. I've ordered a 28BYJ-48 to take apart and may try re-gearing it internally.

If anyone is interested I've had some replies on a stack exchange post with more photos: https://electronics.stackexchange.com/q ... pper-motor