Is it possible to control a Jaguar with a potentiometer?

[DonRotolo]

Originally Posted by dave1027

You should be able to generate the appropriate signal using a basic 555 timer circuit.

I made a post on chief Delphi a few years ago showing a circuit that does exactly that. You might want to search for it.

[Ether]

Originally Posted by DonRotolo

I made a post on chief Delphi a few years ago showing a circuit that does exactly that. You might want to search for it.

I had the link handy so here it is:

http://www.chiefdelphi.com/forums/sh...3&postcount=34

[dtengineering]

I was wondering if it might not be possible to set the Jag to Analog Position Control Mode using an RS-232 or CAN connection, telling it to head for the 1.5V position. Then set only a proportional response, no I or D.

As you swing the pot to one side the motor should move in one direction, with increasing power relative to the "error" in the Pot. Swing it the other way and the motor should turn in the other direction.

Not that there is anything wrong with the other ways, but is there a reason this wouldn't work? I haven't played with the Jags for years, so I'm just working from the manual here...

Jason

P.S. Not that I don't use 555 timer circuits, but I'd probably be tempted to use a PIC or Arduino to generate the serial code for the Jag, but they could also generate PWM code quite easily. A 16f88 or similar would be a one chip solution... easily programmable in Great Cow Graphical Basic, using a PICKIT2 or PICKIT3 programmer and a breadboard.

[roystur44]

Here is the link to our white paper with pics.

http://www.chiefdelphi.com/media/papers/2699

[Hugh Meyer]

Here is a link to what we use:

http://gadgetgangster.com/find-a-pro...projectnum=206

You need to lower the value of the output current limiting resistor then it works great.

I suggest you add a current probe to your test rig so you can measure current while you are testing your systems. A 10 gauge wire 12" long works perfect. Use that wire as a shut, it will measure 1 millivolt for 1 amp. There are several threads here that discuss that scheme. Do a search for stangsense.

Here is a photo of our rig that can control two motors. In this photo the motor is running and drawing 3 amps. By having this information you can know that your motor is running within the correct range and that the magic smoke will remain.

-Hugh

[Gdeaver]

I have found that a Pololu RC controller to be handy for some robot testing.
http://www.pololu.com/catalog/product/1352
You could do the same thing with an arduino, but the control application that you would run on the laptop makes it easy to use. Can also read analog 0 to 5 volts so it can double as a sensor checker.

[anthonyttu]

Using the CAN/RS-232 method only works if always want to have a laptop or MSP430 or something else to always connected to the CAN buss. If you are going that far there is not much to gain.

[MrForbes]

I'm really impressed with how much effort we'll go to, to avoid buying a $25 servo tester.

[DonRotolo]

Originally Posted by dtengineering

A 16f88 or similar would be a one chip solution... easily programmable in Great Cow Graphical Basic, using a PICKIT2 or PICKIT3 programmer and a breadboard.

A 556 also is a one chip solution. For me, a non-programmed solution is almost always superior to one needing programming. (Not that I dislike programming, it just seems like overkill)

Originally Posted by squirrel

I'm really impressed with how much effort we'll go to, to avoid buying a $25 servo tester.

Just like Ham radio operators, robot people are cheapskates. Thus the attraction of building a solution for $4 to avoid spending $25.

[jspatz1]

Do you have access to a Vex controller? Small and portable, remote RC control.

[dtengineering]

Originally Posted by DonRotolo

A 556 also is a one chip solution. For me, a non-programmed solution is almost always superior to one needing programming. (Not that I dislike programming, it just seems like overkill)
Just like Ham radio operators, robot people are cheapskates. Thus the attraction of building a solution for $4 to avoid spending $25.

True, but the PIC wouldn't need external capacitors or resistors.... well, other than the pot.

But I gotta admit... the 556 even with the resistors and the caps would be slightly cheaper than the PIC... much cheaper if you didn't already have some programming hardware around.

I've used both 555's and microcontrollers in projects... I tend to like the microcontroller route because when I screw something up I can usually fix it by upgrading the software. (Why, yes. I do use Windows, why do you ask?)

Jason

[ratdude747]

If it were me I'd just use BDC-Comm and build a CAN to Serial Cable and a CAN terminator and use a laptop (or netbook with a USB to serial adapter).

Then again, I did write a paper on doing such so my opinion is kinda biased?

But really, all one needs is a laptop (with windows and serial means ^ ), a junked serial cable, two RJ12 (or an RJ 12 and an RJ 11), and a 100 ohm resistor. That could be built for say, $5 (or less if you already have CAN parts to your disposal). It's also pretty easy if you know how to use a multimeter and how to solder. The one I built to make my paper took like 10 minutes to make, and that was with pausing to snap pictures.

[anthonyttu]

My quick calculations for somebody to buy the resistors, capacitor and 555 would be $6.30
TI's launch pad $4.30
Laptop CAN method $1.80 (assuming you have a computer)
RC Tx/Rx $36.50 (wireless method for "safe testing")
Jumper wires $0.12

PS you only "need" the terminator for high baud rates. Not critical for testing

[Ether]

Originally Posted by anthonyttu

My quick calculations for somebody to buy the resistors, capacitor and 555 would be $6.30
TI's launch pad $4.30
Laptop CAN method $1.80 (assuming you have a computer)
RC Tx/Rx $36.50 (wireless method for "safe testing")
Jumper wires $0.12

PS you only "need" the terminator for high baud rates. Not critical for testing

You left one out:

RS-232 method: $0.10 for a diode.

[ratdude747]

Originally Posted by anthonyttu

PS you only "need" the terminator for high baud rates. Not critical for testing

Yeah, but as cheap as (and common) as 100 ohm resistors are, why not?

Maybe they changed the firmware (or its a gray jag specific issue), but I somehow I remember back in 2011 (the last time I worked hands on with jags directly) a missing terminator would knock out the last jag in the string (it happened before); the last jag became the terminator. In the case of a solo jag, that meant that the jag wouldn't work at all.