VEXnet Initial Impressions

So, I’ve had my shiny new VEXnet Upgrade Bundle for about a dozen hours now, and so I thought I’d post a few of my initial thoughts and impressions…

The management summary is: It all works as expected, and works well.

Radio Link
Lots of folks seem to get caught up on the fact that 802.11 is being used, but it is best to simply ignore that for now. This is a point-to-point replacement for the crystal radio system, and you don’t have to do anything to manage it from a Wifi perspective; You don’t need a wifi network, but if you have one, no problem.

I didn’t notice any obvious latency in the system; servos and motors seemed to respond instantly to changes on the joysticks. I didn’t spend any time looking at signal strength/range or interference immunity. I was just playing around with the functionality and getting a feel for how it all works.

Installation
Installation is pretty much like it appears in the posted photos. Both the transmitter and receiver attach parasitically to the existing Vex components, so there are several connections to make. Cable lengths are just long enough to fit comfortably, but not so long as to create a cable management issue.

The plastics quality and finish looks just like the rest of the Vex system components. The exception is the 9V backup battery holder, which seems to be a softer plastic than I’ve seem them use, but I think that is necessary so you can get the battery in and out. Everything seems well made and high quality.

Pairing
Once it was all hooked up, I had to pair the transmitter and receiver. This is done by plugging a special USB A<->A cable (included) into the USB ports on each unit. Paring took about 10 seconds. Now that that was done, the receiver and transmitter reliably locked onto each other in less the 10 seconds of powering on every time.

Functionality
My goal for the day was to get through as many of the advertised features as possible. I tried all of the following:
[LIST=0]
*]Single-operator control
*]Dual-operator control
*]Wireless programming (EasyC Pro)
*]Wireless On-line window (EasyC Pro)
*]Wireless PrintToScreen() logging (EasyC Pro)
[/LIST]

Everything worked exactly like you would expect it to; no real surprises to report. It was particularly cool to get log data from a robot that was 10ft away going through its paces, and then download new software, and then watch it go again. All without bending over to pick the robot up!

At no point during any of my tests did I have a control glitch or anything that concerned me. The link seemed to be rock-solid. This testing was done in a residence with a somewhat active wifi network.

Resolution
One test I was eager to try was a signal smoothness test. Basically, I just plugged a servo in to Motor port 1 and slowly (smoothly) moved the Ch1 joystick from side to side. I did this with VEXnet and then with a traditional tether cable, and compared the smoothness of the servo each time.

The VEXnet control seemed much smoother to me than the tethered connection. I don’t have any easy way to quantify my results, but it certainly looked smoother to me. This will probably be an advantage to drivers trying to use finesse in a delicate situation.

Product Design
From a product design perspective, I think IFI did a great job creating these add-ons. But they are add-ons, and as such are not as streamlined as we might like. The feel of the transmitter in your hand is noticeably different, though not so much that it should be a problem. Changing a transmitter battery requires a bit more care to deal with the additional cables and connectors. Again, no real problem, but I do look forward to the more sleek VEX 2.0 hardware with VEXnet built-in.

This product lives up to my expectations, and I’m pretty picky :wink:

Cheers,

  • Dean

Thanks for the input Dean - we haven’t purchased any yet, but the apparently smooth transition is making me feel better already. One question though: is the program loading speed noticeably faster than the old serial connection, or is that restricted by other components?

Looking forward to the wireless programming (wow, is that going to make my life easier)…

John

It felt about the same, but I didn’t time it. I assume the limiting factor is the serial port connection speed (orange cable) which is still used for programming via VEXnet.

Cheers,

  • Dean

When I was helping massey university with their VEXNet system, the download speed was quite a lot slower using wireless (RobotC), eg 5 seconds instead of 1 second, but it was really usefull, especially when you keep having to take out the orange cable to test the robot and have to plug it in again to download a new programme.
I would say that at least 50% of the time it takes to programme a robot is spent on resetting the field, pluggin in the cable, and downloading the programme. VEXNet make this a lot easier.

For RobotC did you have to download the beta version (something like v1.56) to use the wireless programming or will the current version (v1.40) work. I just want to know b/c our team should be receiving the VEXnet upgrade kit in a week or two (even though we ordered it like a month ago).

I only tried EasyC Pro. According to this wiki page, you need 1.54 Beta (download link provided on the page).

I’ve been meaning to play with RobotC anyway, so this is probably a good time to give it a whirl…

Cheers,

  • Dean

The RobotC v1.54 is no longer there… I have v1.52…

Edit: I just checked the v1.54B link again, still not there, and I have asked a month or so ago about the Newer Version of RobotC.

Some other questions are slowly trickling into my head:
[LIST]
*]How does the battery life of the VEXNet system compare with the 75 MHz system? Is there any noticeable difference?
*]All indications I have seen tell me that the old transmitter antennae are no longer needed. Can somebody confirm that this is true?
[/LIST]

Thanks,
John

The Rice University Robotics team had great experiences with the VEXnet system. We ran back to back to the college finals with decent battery life (comparable to the Analog transmitter systems).

Since the communications are digital, the analog hum of the motors disappear and that will reduce current draw.

They made the 1.52 to 1.54 change to RobotC at the VEX world competition, after much discussion with the RobotC developers, I found that the change in 1.54 was really not needed. I would imagine the developers retracted that version after the competition was over.

I can confirm this. I did my testing using a transmitter without an antenna. (The antenna was broken by a student so I removed it).

I believe that plugging a tether cable into a transmitter disables the 75MHz modulator. So not only is the antenna not used, the transmitter isn’t even radiating any energy in the 75MHz range.

Cheers,

  • Dean

This made me curious so I took some readings. I used an adjustable benchtop power supply so I could set the volts and read the amps.

VEX Transmitter tests, run at 9.6VDC:
[LIST]
*]No crystal, no VEXnet: ~50mA
*]With crystal (no VEXnet): ~165mA
*]With VEXnet (no crystal): ~ 235mA
[/LIST]

VEX Microcontroller tests, run at 7.2VDC (no motors or sensors attached):
[LIST]
*]No receiver: ~54mA
*]One crystal receiver (no VEXnet): ~64mA
*]VEXnet receiver (no crystal): ~300mA
[/LIST]

Clearly, the Wifi radios take more power than the 75MHz radios, but I think the amounts are insignificant.

On the transmitter side, the theoretical run time (using a perfect 1000mAH battery) goes from 6 hours to 4.25 hours - not a practical issue as far as I’m concerned.

On the receiver side, the increased power drain will be insignificant compared to the power drawn by the motors, so that is a non-issue. In fact, Lyncas’ point about the analog hum, and the fact that the servos are smoother will probably save way more power than the increase I’ve measured here. You may actually get longer run-times on the 'bot side.

Cheers,

  • Dean