Closed TzOk83 closed 6 years ago
Does it have a green base by any chance?
Yest it has... I've just found another closed issue concerning this one.
Buy another rotary encoder. Maybe I can support this type if I'm sent one. But it's easy enough to find cheap ones that do work.
I just did a search for "green encoder" and didn't find any results? I've been sent some of these from the same seller that was supplying blue ones, I thought they were the same, but I presume from reading this that they're electrically different somehow?
The green base isn't a selling point, it's just an observation that these incompatible encoders seem to all have a green base rather than blue. Maybe the green ones can be supported but I'd need one to test with.
Ok, I'll give one a try on the next set of builds I do, see if I can see a difference in the waveform output. Happy to send you a couple if it helps?
Yes that would be helpful thanks. I guess confirm yours don't work, first.
Even within a single manufacturer's range of encoders you can find models that do one full quadrature cycle per detent and others that do only a quarter cycle. For example most E-Mu synths use a Bourns ECW0D-C24-BD0009 where the D0009 indicates 36 quarter cycle detents but there's also an E0036 variant which has 36 full cycle detents. (see https://www.mouser.co.uk/datasheet/2/54/CW1J-777413.pdf)
I'm guessing somewhat, but your encoder sounds like it's a half cycle per detent, whereas FF is expecting a quarter cycle per detent. Do you have the exact part number?
The trace posted in the second half of ticket #50 seems to show a full cycle per detent, but inconsistent phase difference between the encoder pins. So I'm reluctant to spend time on this unless I receive an encoder to test myself.
EDIT: OTOH the issue in #50 sounds different -- always one direction, or detent missed entirely. Different than this issue or issue #119
Okay could you please try the following firmware, requiring rotary = half
in FF.CFG:
Works perfect!
That's great! I will test on my basic full-cycle encoder tomorrow (I changed round the whole rotary encoder logic to make it more straightforward and logical). If it works okay I will commit the patch to master and it will get rolled into next release 0.9.22a.
@keirf Thanks for implementing this. I bought another encoder meanwhile but it does not fit my 3d printed external case so well as the 'half' cycle encoder I designed it for. I just thought, I'd have to live with that. It's nice to see that I can put the original encoder back now! By the way, the new encoder also has a green base, but did just work with the 'simple' setting. So, there seems to be a lot of scattering in this area...
Hi.
Yes, I the green encoders that I was sent seem to work in exactly the same way as the blue base ones, so not sure what’s going on with these el-cheapo suppliers, but either way the firmware works. I think it’s even better responsiveness now as well.
Thanks, Mark.
From: Torsten Kurbad Sent: Sunday, June 17, 2018 8:16 PM To: keirf/FlashFloppy Cc: RetroClinic ; Comment Subject: Re: [keirf/FlashFloppy] Rotary encoder - either requires 2 "clicks" or skips by 2 positions (#130)
@keirf Thanks for implementing this. I bought another encoder meanwhile but it does not fit my 3d printed external case so well as the 'half' cycle encoder I designed it for. I just thought, I'd have to live with that. It's nice to see that I can put the original encoder back now! By the way, the new encoder also has a green base, but did just work with the 'simple' setting. So, there seems to be a lot of scattering in this area...
— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.
Maybe it is my specific issue, but I've got rotary encoder (EC12 clone, 30 imp.), which either moves 1 position, every two clicks (simple), or skips by 2 positions (Gray). Actually - when Gray pattern is chosen it moves 1 position every 1/2 click - if I stop in between clicks it moves only 1 position, but when I allow it to click, it moves to the next position.