Author Topic: robotic potentiometers  (Read 27542 times)

~arph

Re: robotic potentiometers
« Reply #80 on: July 21, 2011, 08:19:32 AM »
there come the led indicators..

rip

Re: robotic potentiometers
« Reply #81 on: July 21, 2011, 11:44:31 AM »
Yes, but fidelity may be a problem as well with leds. Sure the digital pots would provide good grandularity, but then you must deal with zipper noise with the IC.

Earilier, I too was thinking a rotary encoder paired with a led ring would work, but I believe a LCD screen displaying actual value would be better. Ultimately, though the mechanical knob turning provides a more user friendly feedback.

I still believe the stepper motor is the best solution.

A significant problem that I don't believe has been addressed here is the layout for such an amplifier design. The x99 was only a preamp, to house the power amp and all the transformers with the stepper motors is no easy task. Sufficient room and shielding must be accounted for.

For a DIYer using COTS (arduinos, and easydrivers, etc.)  means a system that would ultimately provide a solution with un-neccesary un-used space because each of the COTS are not specifically designed for the amp application.

Rip


tokyoburns

Re: robotic potentiometers
« Reply #82 on: July 25, 2011, 08:55:11 PM »
@RIP
I am not building an amp. This is a design for a pedal.

 I have been working on this design for a long time. I am not a born tinkerer like other people on here. A lot of my time is going in to research. I had to teach myself how to program which im still learning how to do. I am still saving up but my next step is to purchase the motors, pots, pic chips, pic programmer, parts for stepper driver circuit and so forth. I have a good list put together of items i found on ebay that will seem to work.

motors - http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=220520400102&ssPageName=STRK:MESINDXX:IT#ht_981wt_1054

programmer - http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=220623060314&ssPageName=STRK:MESINDXX:IT#ht_1079wt_1054

chip - 16f88

motor clutch - http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=220676556168&ssPageName=STRK:MESINDXX:IT#ht_2569wt_1015

what do you guys think? The motors are the closest thing I could find to what I need. It will allow me to get 150 different positions on a 270 degree pot so I think that should be pretty good. this with a few other caps and resistors should be enough to test my stepper programming.
expansive

Gurner

Re: robotic potentiometers
« Reply #83 on: July 26, 2011, 01:01:06 PM »
Have you confirmed with the seller the resolution of those steppers? (there's no mention in the advert....normally a bad sign!)...some smaller steppers aren't 200 steps per rev.

R.G.

Re: robotic potentiometers
« Reply #84 on: July 26, 2011, 01:14:48 PM »
I'm with Gurner - I wouldn't buy the steppers unless I got some information that they had lots of steps. The seller should be able to guarantee that they have many steps or to tell you the maker's model number so you can look it up.

The shafts are 5mm and 4mm diameter; so the clutches you're after won't fit the shafts directly, nor will a standard knob, which are most often for 1/4"/6.35mm diameter pot shafts.

I personally would bag the clutches unless you just like the whizzo mechanical look. 1/4" fuel hose from a car parts place will slip right over 1/4" shafts (do you detect a bias in my advice toward getting 1/4" shafts?  :icon_biggrin:  ) and be held in place fine with screw-type hose clamps, in addition to being much cheaper.

The programmer is OK, I guess. I wouldn't go with one of the 16F8x series. The 16F6xx series is newer, cheaper per function, and in general better.

R.G.

Quick IQ Test: If anyone in a governmental position suspected that YOU had top-secret information on YOUR computer, how many minutes would you remain outside a jail cell?

rip

Re: robotic potentiometers
« Reply #85 on: July 26, 2011, 03:52:37 PM »
I agree about the clutch, fancy but not needed. If you got the money fine but it's was meant for 1/4 shafts. I would suggest a coupler, get a machine shop if you need to, to make a shaft with a 5mm hole and a grub screw, the other end I would machine a interference fit or close enough fit with may be some weak loctite.

RG the nema 17? Stepper motors are typically 5 mm and those are really the closest ones to a 1/4 shaft I believe the x99 used the same. So I don't think we'll get away from that ;(. Wish there were more choices.

Oh and why not go with the Arduino? Huge following where you will more easily find help to debug

Rip

R.G.

Re: robotic potentiometers
« Reply #86 on: July 26, 2011, 04:41:39 PM »
RG the nema 17? Stepper motors are typically 5 mm and those are really the closest ones to a 1/4 shaft I believe the x99 used the same. So I don't think we'll get away from that ;(. Wish there were more choices.
The old full-high 5.25" diskette drive (from when dinosaurs roamed the world) each had a 200 step, double-ended 1/4" shaft, 5V stepper motor in them. At one point people would pay you to haul them away, but I suspect the vast majority are now in landfills. Too bad.

Using rubber hose for a couple would make you able to squash it down from 0.250 (1/4") to 0.200" (5mm, more or less). Knobs are harder.

I have a lathe, but most people don't.
R.G.

Quick IQ Test: If anyone in a governmental position suspected that YOU had top-secret information on YOUR computer, how many minutes would you remain outside a jail cell?

tokyoburns

Re: robotic potentiometers
« Reply #87 on: July 26, 2011, 09:24:54 PM »
If you look at the pictures from the motor there is one that has information about the number of steps. 3.6 degrees 100 steps/revolution. I wrote my programming for half steps so that gives me a 1.8 degree angle and 150 different positions across a 270 degree pot. The coupler is just one that I was looking at. I thought it looked pretty spiffy but I wasn't married to it. I'm sure I can find one with the right dimensions. I am not using regular knobs for this. they are going to be handcrafted per the aesthetic design which I'm keeping to myself. So I am not worried about about using common dimensions. It has been very difficult to find the right motor at the right price. These are definitely the best ones I could find. there was a different motor listed on an obscure sight that i didnt feel all that comfortable about getting over the net.

http://eolsurplus.com/StepperAndServoMotors.html

"Sigma Instruments, Inc. 20-2220D-24435, $20.00, 2 Available
Stepper Motor, 200 Steps/Rev, 50 oz/in, dual 1/4" shafts, NEMA 23 single stack, 4 wire."

If I could find these somewhere else they would be my first choice.

I don't really know much about the arduino I have already spent a lot of time learning how to program a PIC so I think I will stick with that. Like I said I have most of the programming written and the program im using "flowcode" lets me test out my programming on virtual components including a visual of watching a custom stepper spin. Not that I dont count on there being bugs to work out. But I think Im already at a really good place with the PIC. That being said...

@RG

If you think the 16fxx series are better I'd take your word for it. I'll give it some tests on the program i mentioned earlier and see if it works out.

One thing I haven't figured out yet is how to display numbers on a 7 seg LED. It seems to me that I'd be able to just apply voltage to whatever leg i needed lit to make a number but it doesn't work on my virtual testing. There are preset macros to call up letters but not for numbers so I guess ill just have to experiment with it when I start putting it together.

Did I mention I started a product design in google sketchup. It didnt come out the way i wanted it to but at least i started to learn how to use that program and came up with something similar to what it is supposed to look like.



expansive

R.G.

Re: robotic potentiometers
« Reply #88 on: July 27, 2011, 11:11:24 PM »
I have already spent a lot of time learning how to program a PIC so I think I will stick with that. Like I said I have most of the programming written and the program im using "flowcode" lets me test out my programming on virtual components including a visual of watching a custom stepper spin. Not that I dont count on there being bugs to work out. But I think Im already at a really good place with the PIC. That being said...
You gotta program with what you feel comfortable with, and what you have the tools for. I've written code in Fortran, Pascal, COBOL, ALGOL, APL, BASIC, 360/370/390 assembler, 6502 assembler, 8051 assembler, 68000 assembler, X86 assembler (d@MN the byte reversals), several RISC architectures, C, C++, and a few others.  I'm kind of at the point where I don't much care about the language. I write down what I want it to do, then go translate my design into the language of the day.
Quote
If you think the 16fxx series are better I'd take your word for it. I'll give it some tests on the program i mentioned earlier and see if it works out.
Again, go with what you like. I like the 16F6xx right now. But I used the 16F54 back in the day.

Quote
One thing I haven't figured out yet is how to display numbers on a 7 seg LED. It seems to me that I'd be able to just apply voltage to whatever leg i needed lit to make a number but it doesn't work on my virtual testing. There are preset macros to call up letters but not for numbers so I guess ill just have to experiment with it when I start putting it together.
Seven segment LEDs are just an array of LEDs. You have to put a *current limited* voltage across the segment, and in software translate the number you want to show to the set of segments that make that character. This is almost universally done by table lookup. Google "seven segment display". The segments are given standard names of "a, b, .., g" and "dp" for decimal point. You will have to know whether your display is common cathode (all the LED cathodes go to one pin) or common anode (all the anodes go to one pin), other wise the LEDs get wrong polarity voltages and won't light up. They are LEDs, but they are still diodes.  :icon_biggrin:





R.G.

Quick IQ Test: If anyone in a governmental position suspected that YOU had top-secret information on YOUR computer, how many minutes would you remain outside a jail cell?