Mounting of router

[goofy173]

I just looked at the 2014 FRC Inspection List and under Robot Radio it says, "Radio must be mounted so that its LEDs are visible to field personnel" and referencing R62 The Wireless Bridge must be mounted on the ROBOT such that the diagnostic lights are visible to ARENA personnel.

Well, our electronics board is installed with drawer sliders and is located under the robot deck. My thinking is that the "radio" router is not going to be allowed underneath there. We were only worried about inspection where we could roll the board out on the drawer slides.

I think if you are at the right angle, you might be able to see the lights, but without being next to the robot until tomorrow, I won't know.

So I guess we'll have to mount it up on the deck somewhere or am I misunderstanding it?

Thanks for any input. If I don't respond it's because the internet at the school won't allow me to post here, but I can read everything. Weird but that's the way it is.

[eedoga]

Couple of things. For the Arena personnel being able to see your router makes it easier to diagnose why your robot won't connect to the field. Also, we in the past have made the mistake of mounting our router low down in our robot...Invariably this made it so that our robot would not connect to the field, or would connect, only to lose connection mid match. The robot itself provides interference making it difficult to maintain a connection, especially in a noisy environment. As a result we decided to always mount the radio higher up on the robot and out in the open if at all possible.

Hope this helps.

Edoga

[Jon Stratis]

Visibility of the router can only be for your benefit - the field personnel use it to help diagnose issues with your robot and team.

Additionally, it's a good idea to mount your router high up, away from metal and away from RF-generating sources (like the motors). Having a router buried inside a robot makes it much more likely that you'll lose connection during a match!

[TogetherSword8]

Is there any reason why you wouldn't be able to mount the router close to any other electronics, such as Crio and Digital Sidecar? Just double checking because we normally do this, and we have a track record of having more connection problems than other teams...

[Alan Anderson]

Keep the router away from electrically noisy components. That includes the Power Distribution Board, the cRIO, the Digital Sidecar, the 12v-to-5v converter, motor speed controllers, and ethernet-to-can gateways.

If you regularly mount the D-Link near other electronics, and you regularly have connection issues, take it as an indication that you should stop doing it.

[goofy173]

I wondered earlier if there could be a problem with the router and the network cable near motor controllers. I know if it was a big A/C freq drive like I work with, it could be a problem, but didn't really consider it much with these small DC drives.

Well, being that our router will now have to come out of the basement it will be away from most things. I'll have to tell them to not put it near a motor.

I also hadn't thought about connection issues. That's a good point.

Thanks for the responses.

[philso]

I wondered earlier if there could be a problem with the router and the network cable near motor controllers. I know if it was a big A/C freq drive like I work with, it could be a problem, but didn't really consider it much with these small DC drives.

Well, being that our router will now have to come out of the basement it will be away from most things. I'll have to tell them to not put it near a motor.

I also hadn't thought about connection issues. That's a good point.

Thanks for the responses.


From having spent many hours in an EMC test lab, I definitely recommend you take Alan's advice. I sometimes view the AC motor drives I do R&D on (88-2000 hp) as noise generators that are their own worst enemy.

[DonRotolo]

Remember folks, it is still a radio. The antenna needs to be as far away from any other metal as you can reasonably make it.

[goofy173]

I made my feelings known today, and to no avail, so if there is a problem, I'll stand back and say, "Told ya so" and go from there.

The lead mentor believes that as long as the field person can come out to the robot and see the router lights without too much trouble, we'll be OK. I'm going to have the kids mount the router at an angle so that it can be seen looking down into the front, just above the bumpers. If inspection says something about that, then again I can say, "Told ya so."

I am fighting the shop right now because they wanted the electonics on a drawer, and both times it was test fit, it didn't. The first time they forgot to take into account the height of the cRIO. This time new items they mounted on it blocked the drawer from even starting to slide in, so they rotated the drawer 180° and although it still didn't fit, they want that. So now all the electronics have to be removed and installed differently, and where I was told we could run wires from side to side, is now not available for wires. All this and they expected a running robot today. Very frustrating for me being a rookie mentor.

I just told the kid in my electronics team that I will be there when they try and blame us for any problem or being late, and I will remember that the shop changed everything for us twice (so far). The mentor in charge of the build (also a rookie) complains about a lack of communication between teams. I guess he feels that being in charge of the build means he doesn't have to communicate with anyone.

After 27 years of being in a union shop, I'm used to getting blamed, so its no big deal to me at all. I'm just going to go to bat for my students because they've done everything they've been asked to do.

[philso]

Sorry to hear about your trouble with "office politics". Your students are getting an early look at what they are likely to have to face in the future. Hopefully, your students will understand why you are telling them to do certain things like "communicate frequently and clearly with the other departments".

[DonRotolo]

If it were me, I'd be sure to print up a few cards that say "I told you so!", so when it happens at inspection (and it will), you can hand them out.

To help your kids manage the stress when they fail inspection, have them think of exactly how they are going to fix it, and make up all (or many of) the wires, brackets, and things they'll need to do it in the pits.

[yash101]

Try to have it visible. It makes it easy to debug what is going on. Also, make sure you can tether to the bot easily. It's very annoying to lift the robot and worry about the wire getting stuck in the wheels to just tether to it!

Also, KEEP IT AWAY FROM THE CONVERTER!!!
That tiny box is actually a tiny box with a big potential of interference!

[Alan Anderson]

The lead mentor believes that as long as the field person can come out to the robot and see the router lights without too much trouble, we'll be OK.

Please let the lead mentor know that many people with years of experience as FRC technical advisors say to put the wireless bridge out in the open. Not just because it keeps the lights visible, but because it is a radio transceiver. If it's surrounded by enough metal and electrical components to make it hard for a person to see, it'll very likely be hard for the field's wireless access point to "see" it too.

[goofy173]

I said to him, "So we to are ignore those that posted that they've had problems?" His response was that it's hardly metal 'cause it's aluminum. I believe he has expertise in networking, but I honestly don't know how much.

I've had an offer for me to take a pic tonight and it be shown to a chief inspector and at least I'll know if its ok to be almost hidden.

I'm also having the kids leave enough power wire, held by a tie wrap, so that it can quickly be mounted on the deck. I asked for a hole thru the wooden deck in case, but I know that will be ignore.

I don't push my knowledge and electronic degree on the other mentors if they decide that things are to be done a certain way. I figure they should know more about building the robot and the rules than me. I think new blood is not always appreciated.

[Conor Ryan]

I know the following is likely to be lost in the bowels of CD:

While we are on this subject, try to mount your Radio "Pancake Style" to maximize the use of the two antenna's in the router with the field access point. Ideally mount it about at least 18" off of the ground.

You will maintain a higher quality connection if you do this. You do not have to do this, it is a best practice.



As for making the router easy to find, my colleagues around the field will thank you. We will further thank you if you make it easy to observe the status lights on the CRIO, although we do realize you can just follow the ethernet cable out of the back of the router. Making our life easier is like giving us cookies.



Well, our electronics board is installed with drawer sliders and is located under the robot deck.

Drawer Slides: Did your team realize that if you use drawer slides, you can't use the "slide" feature after it is all wired? I learned that my rookie year the hard way. Back in the IFI days when we used drill motors as the kit drive.

[Al Skierkiewicz]

OK,
When trying to impart the wisdom of others, one thing to point to is the signature of guys like Don and I. I am a broadcast engineer and Extra Class Ham Radio operator. Don is a Ham with significant digital experience. Alan is a trusted network and software specialist with a lot of wireless experience.
Aluminum at 5 GHz might as well be lead.
For everyone, the DAP-1522 has two antennas mounted on the board. They are verticaly polarized with respect to the bottom of the AP/Router. That means they are most sensitive to radiation out to the sides, front and back of the radio. The circuit board is a ground plane effectively blocking all reception in that direction (out the bottom of the radio). The nature of 802.11N is that both antennas are required for max bandwidth to be achieved. Both are used for receive and transmit. In addition to the shielding effect provided by metallic objects, the metal serves to de-tune the antennas making them less effective. This effect, of course, is distance dependent. As Alan has pointed out, there is some significant noise generators on the robot. Of those that are the worst for interference is open frame motors (those that you can actually see sparks through the openings in the sides of the motors), the five volt convertor power supply, and the supplies in both the PD and the DSC. These noise sources not only interfere with radio reception, they also impart noise directly to the electronics on the board.
Having the radio mounted in the robot such that is "visible" to the antennas on the field, it is not mounted near a noise source and it is not surrounded by metal of any kind is the best of all worlds.
One word on "visibility", inspectors want to be able to help you and insure that you will play all matches at every event. If your mounting is questionable it may pass, but if you want help when you fail on the field, you have to meet us halfway. If we can't see the lights from several feet away when you are dead in the middle of the field, we have to rely on the field logs only, to try and figure out what went wrong. After repeated connection issues or a string of bad packet events we will ask you to move the radio to help you improve.

[Lightfoot26]

OK,
When trying to impart the wisdom of others, one thing to point to is the signature of guys like Don and I. I am a broadcast engineer and Extra Class Ham Radio operator. Don is a Ham with significant digital experience. Alan is a trusted network and software specialist with a lot of wireless experience.
Aluminum at 5 GHz might as well be lead.
For everyone, the DAP-1522 has two antennas mounted on the board. They are verticaly polarized with respect to the bottom of the AP/Router. That means they are most sensitive to radiation out to the sides, front and back of the radio. The circuit board is a ground plane effectively blocking all reception in that direction (out the bottom of the radio). The nature of 802.11N is that both antennas are required for max bandwidth to be achieved. Both are used for receive and transmit. In addition to the shielding effect provided by metallic objects, the metal serves to de-tune the antennas making them less effective. This effect, of course, is distance dependent. As Alan has pointed out, there is some significant noise generators on the robot. Of those that are the worst for interference is open frame motors (those that you can actually see sparks through the openings in the sides of the motors), the five volt convertor power supply, and the supplies in both the PD and the DSC. These noise sources not only interfere with radio reception, they also impart noise directly to the electronics on the board.
Having the radio mounted in the robot such that is "visible" to the antennas on the field, it is not mounted near a noise source and it is not surrounded by metal of any kind is the best of all worlds.
One word on "visibility", inspectors want to be able to help you and insure that you will play all matches at every event. If your mounting is questionable it may pass, but if you want help when you fail on the field, you have to meet us halfway. If we can't see the lights from several feet away when you are dead in the middle of the field, we have to rely on the field logs only, to try and figure out what went wrong. After repeated connection issues or a string of bad packet events we will ask you to move the radio to help you improve.

Thanks Al! This is extremely informative. I learned my lesson when we have had radio problems in the past, and ended up fixing the issues by simply moving it. I try to stress this to my current students, I will definitely pass this on to them! It's a good explanation. ... and since your a HAM, I'll say 73 AL! Thanks! :)

[Al Skierkiewicz]

Seth,
If you want me to talk to your students some time, just ask. See you soon.
Al

[Alan Anderson]

I am a broadcast engineer and Extra Class Ham Radio operator. Don is a Ham with significant digital experience. Alan is a trusted network and software specialist with a lot of wireless experience.

I would like to consider the possibility of setting up a semi-official field-day-like station at FRC competitions. Who wouldn't want a Worked All Regionals certificate? :cool:

DE WB9RUF

[Mark Holschuh]

Thanks to Don, Al and Alan for your wisdom here. In your experience, does the signal light impart any interference with the radio signals? We are currently planning to mount them near each other so that the field crew can find them easier.

[Lightfoot26]

I would like to consider the possibility of setting up a semi-official field-day-like station at FRC competitions. Who wouldn't want a Worked All Regionals certificate? :cool:

DE WB9RUF

Send out some really cool FRC Regional Event QSL Cards! That'd be so cool! haha :cool:

anyway... sorry, I digress!

[Al Skierkiewicz]

The RSL has no real generator in it. It merely is a high brilliance LED. It contains a flasher circuit that is not used in our application. Alan, you never told me you were a ham. A demo table might be a cool thing at events.

[DonRotolo]

I agree, the RSL is pretty quiet. No worries if it's mounted with the radio.

Hey Al, I also was a TV engineer for 3 years...but on the consumer electronics side of things. Now it's automotive.

I also never knew Alan was a ham, but it does not surprise me.

[goofy173]

Drawer Slides: Did your team realize that if you use drawer slides, you can't use the "slide" feature after it is all wired? I learned that my rookie year the hard way. Back in the IFI days when we used drill motors as the kit drive.

Please elaborate. You are saying after we have it wired, we cannot slide out the drawer for some reason? We're using connectors so we can easily disconnect sensors, solenoids and motors at the top of the deck, and then the drawer can slide out.

[Ikillee]

Can the router be placed on top of the battery? That's the best place we got where the router is the least interfered with. Other places are underneath our polycarbonate in the middle of all the electronics. My team doesn't want to agree with mounting on top of the battery because the battery could overheat or something.

[Al Skierkiewicz]

Ikillee,
During match play, the battery might get hotter than normal. It also contains a lot of metal. I would think you could find a better place. If it is on top of the battery, don't you have to remove it to change batteries? You don't charge batteries in the robot do you?

[Ikillee]

What if it was mounted sideways right beside the batteries?

[yash101]

Here's what I believe is the best:
Motors are probably one of the worst interference, even the smallest ones. Motors use electromagnets to convert the electrical energy to kinetic motion. Motors, because of that, when in close proximity of the bridge, can cause severe problems. That radio wave will interact and deform the radio wave of the bridge, causing it to increase it's power until it's max threshold. Then, you will notice your connection continuously dropping. When you go full speed forward, you lose connection because there is a sudden load of interference. It isn't because of the voltage drop because there is a buck converter, probably containing some smoothening capacitors. What the bridge is encased in also maters a lot. Start a router and keep the device next to it. See how good your signal is. Then, go to the other side of a brick wall, and voila! You will notice an extraneous drop in signal quality. In my room, where I keep my laptop, if I keep it 3 feet higher, on my desk, I'll get a good signal. However, when I put it on the floor next to a bunch of transformers, I am lucky to get 3 bars of signal. Oh, remember that I'm running a hacked router with DD-WRT, so I can change the transmission power manually and max it out to 251mW, powerful enough to give me a .25 mile range with another DD-WRT router at the same transmission power! Also, keep the converter away from the bridge. It constantly changes in frequency so it is almost impossible to go against! As Al said, try to have your bridge high up and flat.

This year, our bridge is on a tall pole, distant from ANY radiation, except the AXIS camera. Otherwise, it is surrounded with pneumatics and a compressor 3 feet away. That aught to give a decent connection!

[Al Skierkiewicz]

Ikillee,
Same but different, the battery contains a lot of lead plates and connecting bars.
Yash,
Not exactly. The motors can produce interference due to the sparking that is normal in DC motor brushes and commutators. Think spark gap transmitters. The closed motors tend to reduce this spark interference but RF does stray along the motor leads as it leaves the motor case. Twisting the motor leads can help somewhat. Your results will vary.
The effect you experience with a brick wall may be more due to the materials inside the wall than to the wall itself (Pipes, conduit, steel structural materials, wire). If that wall has plaster on the surface for instance, it is likely to have a steel mesh screen (lath) to help the plaster stick to the wall. The convertor makes a lot of RF noise because of the way it switches current to the internal transformer. This produces significant, high slew rate, pulses with lot's of ringing. While the frequency of operation is relatively low, (likely around 100 kHz) it is the ringing that is much higher in frequency and that which causes the interference. when placed close to the bridge, the comparative signal strength from the convertor is orders of magnitude higher than that received from the other radio. This tends to fall off rapidly with distance (following the inverse square law).
I would be remiss if I didn't say something about exposure here. The radio output from wireless home routers is intended to keep exposure limits low and maintain FCC specifications. Please consider what you might be exposing yourself to when making power adjustments to a product of this type, particularly if you spend a great deal of time in one location near the source.

[goofy173]

Well, if I have my way, I will be making a change to it today. The mentor that was against moving the router up and quit Saturday! He's our head mentor. Kind of weird to quit just before build is over. I won't have time to get it out of the drawer but if we get a chance before bagging it, I'm going to put an angled piece of plexiglass aiming it up at about 40°. The mentor that quit was fine that the lights only showed thru the heavily smoked Plexiglas, which I think would have been behind the bumpers anyway.

If worse comes to worse, 1 hole and some Velcro on the deck and we can mount it up there at competition.

[philso]

It looks like Fate has intervened on your behalf.

Please use polycarbonate (Lexan, Macrolon, etc.) rather than acrylic (Plexyglass). The acrylic sheet will shatter easily if stressed and especially if it is subjected to shock (there are other threads here about this). Polycarbonate is very resistant to shock and in the rare instance when it does crack, the cracks propagate slowly. You can get polycarbonate sheets from your local Home Depot or Lowes store.

Velcro and zip-ties are all you need to hold the router onto your polycarb plate since it is so light.

[Al Skierkiewicz]

A rule of thumb I was taught for low frequency antennas separation should be a minimum of 1 wavelength and 10 wavelengths or more was ideal. (for non resonant metal objects) That will at least give the least amount of detuning of the antennas. At 5.4 GHz the wavelength is about 2.6 inches. So I would recommend that both sides of the radio have no metal within two inches. The antennas are almost centered in the box so they are about two inches from the front and back of the case.