PC control of a Victor884/883

I’m working on a project that will require me to control several Victors with a PC parrellel port. I was thinking that this will be easy to do but I have a few questions. What is the purpose of all three connectors on a PWM cable, are only the ground and postive connectors used? Aslo, what is the range of duty cycles the Victor 883 and 884 can handle? And finally how much time does one cycle take?

Thanks in advance.

I don’t have experience with a victor but other similar speed controls I’ve used like to take pulses between 1.0 and 2.0 milliseconds in length with a 1.5 ms pulse being the neutral throttle posititon. These pulses should occur every 20 ms.

I have an osciliscope at my displosal, could I use it to find the length of the pulses the RC sends to a Victor? However, I have never used an osciliscope other than just playing around. How can I use it to measure length of the pulses?

The red and black wires on the PWM cables are +7.2 volts and ground respectively. The white wire is what you use to vary the pulse length.

This Kit shows how to access a printer port via the various languages. I can vouch for the I/O box too. Very impressed. Using this box and the example code, you should be able to control a Victor tolerably well. I would not do anything safety critical because I would guess it could freak out when the PC gets too busy or Windows gets flakey.

If you REALLY want to control a Victor using a PC, I would recommend either THISfrom Scott Edwards which uses an RS-232 port and controls up 16 servos (8 per board, 2 boards per USB) or even better THISfrom Parrallax Inc. which uses USB to control 16 servos on one board.

I have used the Scott Edwards board and vouch for it. Old FIRSTer may know that FIRST used Scott’s SSC code to run a PIC that was inside the old Stamp2 based controllers (why this explains the 127=OFF set up by the way it is a legacy from Scotts boards).

I have the Parallax board on order (it hopefully ships Friday). I have high hopes for it. By the way, do not be fooled, the Parrallax Serial Servo Controller is not the same as the Scott Edwards board. The Parallax Serial board is essential made to interface to a STAMP2. It requires other glue to interface to a PC RS-232 port.

Good luck.

Joe J.

Wow, thanks for the link. That board is incredible, and incredibly cheap (which is probably why it is backordered). The manual isn’t yet available for donwload. Can you tell me some more about it? Does it plug right into the USB port on a PC? If so, what (program) controls it? Is there software that runs on windows to control all the servos?

You could make a really walking robot with that. What do you plan to do with it?

It is not yet in from their (Asian) manufacturing source. Their tech support claim production delays are the issue not too many orders.

I THINK that the USB will make a virtual COM port that you can access via whatever you want. BUT… …I too am not clear on this. I am inferring much from THISpage (and so experience I have with USB Bluetooth Dongles). Look at this picture that shows and USB to COM port chip and then look at this picture of the USB Servo Controller. Suspicious no?

Stay tuned…

Joe J.

So the yellow signal cable is unused? I suspected this.

Oh. I should have mention that this project is to show the power of Linux, so I SHOULD have little to worry about the OS being unstable. However the operating conditions could get kinda rough. :smiley: :cool: 100 degree Virginia days are not exactly the best operating conditions for a PC, although the water cooling system should take care of this.

Due to the fact that I am using Linux I think it will be easier just to use a printer port to control the Victors, as the USB kits do not seem to offer Linux drivers(no surprise there). From what I have read a parrallel port should allow me to control 5-8 servos or Victors, is this correct?


The yellow and white cables on ‘pwm’ wires are the same. Some cables happen to use yellow for signal, others use white. The signal carried is the same.

-Andy A.

A lot depends on if you are willing to use extra glue chips to make this work. With a serial in, parallel out latch, you could essentially shift in as many bits as you like and controll a TON of servos. But… …if you are trying to more or less plumb a DB-25 connector to a printer port, you’ve got 6 control lines and 8 data lines.

The control lines are STROBE, AUTOFD, SLCTIN,PE & SLCT, but looking at the data sheet, PE, SLCT & ACK are inputs. SO…

I think you are limited to 11 Servos max just based on hardware. With a stable OS and some flexibility to set priorities of tasks you should be in good shape controlling 11 servos. The timing is really not that critical compared to other tasks.

Oh yeah, I would not be so sure about the lack of a Linux driver. See the quote below from here:

These chips have proven their abilities under multiple operating systems with FTDI’s free, top-notch virtual COM port drivers for PCs, Macs, Linux and CE operating systems. The virtual COM port drivers interface just like a serial port with extended data rates. FTDI chips are state-machine based and require no firmware. They can access an optional external EEPROM upon power-up which contains your own USB vendor and product ID should you require this feature.
It seems to me that the FTDI chip that, based on the picture of the board, is used on the Parallax USB servo controller is THE thing that turns a standard Parallax Serial controller into a USB Serial controller.

If I am right, then the FTDI just makes the USB port generate a virtual COM port. AND FTDI claims (above) that they have FREE (as in “free beer” for you Open Source folks reading this) drivers for “multiple operating systems” including Linux.

Look at these two boards:

The only difference really is that the one on the right (the USB one) has the FDTI chip on it. I am pretty sure the the differnence is that one gets it’s serial data directly an the other gets it from the FDTI chip. In that case, a Linux driver that gets you a “virtual COM” essentially IS a Linux driver for the whole kit and kaboodle.

But… …as usual, I could be wrong.

Wrong or no, Good luck. Let us know how it turns out.

Joe J.

I think I will buy one of the Pallax boards, they seem like a good solution. Thanx for the help, I will keep you all posted when I start getting parts in.