Attached is a new schematic for the IR board. I will be posting the layout within the next few days. I am posting the schematic to get feedback from teams before I do a layout. Once the layout is done, I will post the Eagle and gerber files for any team that wishes to build there own board to replace the board provided in the KOP. I have added a diode for reverse polarity protection and an electrolytic cap to help with any ripple. I have also increased the value of the current limiting resistors for the LEDS from 240 to 330. Any other suggestions let me know. Also, the finish board will have a ground plane covering the bottom layer.
IR_BOARD.pdf (142 KB)
IR_BOARD.pdf (142 KB)
Also let me know if I made any mistakes. I was multi-tasking when I made this schematics.
Please don’t be offended, your efforts are truly appreciated, but there is an even simpler, more reliable option available. It is comprised of a single VISHAY IR receiver, a PWM cable and a PWM connector. That’s it!
You then just add a small amount of code provided by Kevin Watson and BINGO, the RC is now your IR receiver as well. You do not need to “train” the receiver, you just pick which four commands you want to respond to in your code.
Details can be found in this thread.
Mike,
Your schematic looks OK to me. One thing to check is the pinout of the voltage regulator. The pins shown look correct for the 78L05, but the part is called 7805. They are different. I would use a 7805, since they are available at Radio Shack. Also, you show an 8-pin dual row header, but the std size is 10.
Concerning the sw-only solution using just an IR receiver. That software seems to be only looking for Sony IR codes. How will that work with a bunch of teams at once? If everybody selects Sony, things will get very interesting.
Good luck,
Bob GriebYou should make sure teams know what the valid input voltage will be with the added diode drop. Will they be able to use the ~7 volt PWM outputs? How about using a LDO regulator? Perhaps eliminate the regulator all together and have a pinout that allows teams to use the RC’s 5 volt output?
Um, are you saying that you can somehow remove the interference in software or by choosing a different protocol? My suspicion is that there is zero difference between the software only solution and the IR board solution. Both are subject to interference because they both use the same Vishay IR detector and both ultimately do the protocol decoding in (PIC) software.
-Kevin
I think the implication is that there is a greater possibility of one team’s code matching another team’s (and therefore causing unintended operations), as opposed to the interference issue which is a physical phenomenon and on which you are absolutely correct.
Just a different point of view.
Don
Mike,
Kudo’s for coming up with 2008 IR Board #2 design. Further recommendation deals with JP1. Since the standard method of hooking up is to use 3 wire, pwm cables, I recommend modifying JP1 to become 4 sets of 3 wires to mate with the pwm cables for the 4 signal input/outputs. The 10 pin JP1 would then become a 3 column by 4 rows or 12 pin connector.
Since your design features a reverse polarity diode without a LDO regulator, I would propose moving the power/ground
input to a separate connector (JP3?). I added this in conjunction with one of those 2 pin terminal blocks to provide power
to one of the KOP IR boards. I am assuming a connection to the 12V fuse panel in this configuration.
I would also consider, if space permits, having open solder pads connected to the 12 JP1 connections to allow soldering a pwm wire with it’s one connector cut off. This would eliminate the possibility of a pwm connector coming lose for team’s that use the solder route.
Can you also space the 78L05 TO-92 pads to be TO-220 pad spacing? This would facilitate replacement of the 78L05 with 7805 TO-220 parts. I needed help from Frank Leritz, HDC-Troy, to swap out the part because the TO-92 pads were so tightly spaced.
With all the people that can possibly respond to this, I believe your opinion carries the most weight considering the amount of time you put into the IR side of this competition. As you well know, the Sony as SIRC protocol is fairly well defined, and it is a very popular protocol.
After playing with Kevin’s software for just a few minutes and one of my universal remotes, I have found a quark and the protocol. We fully intend to take advantage of this quark.
In addition to taking advantage of this quark, there are other things to consider as far as strategy with this game. When and how you use your IR remote will have a big impact on how well the receiver operates in this environment. If the intent is to use the IR during a time period when most teams will not be, and from a location that may not be common, as in immediately after the ball is placed and before the bots are put in to hybrid mode, from the Robocoach Station right next to the Alliance station, then the outcome should be quite favorable.
Your point about the amount of IR signals flying about is quite valid and we have considered that extensively before coming up with our strategy to play this game.
The bottom line is we may choose to use your IR board, or the software only solution. We will determine this during practice on the first day of competition. Personally, I think we may be leaning toward the software only solution because it allows us to place the IR receiver and a much more favorable position on a robot than we had originally placed the IR board. That alone may be the deciding factor for our team.
Ugh, hopefully you mean quirk with the protocol and not quirk with the software <grin>. BTW, I know I owe you an e-mail and I’ll get to it tomorrow at some point.
-Kevin
The purpose of this post was not to provide an alternative but to provide teams that have already implemented the IR board with a replacement if there board was to go south. Using the RC is an obvious option. This is simply a replacement for teams that find themselves in need of another board when one is not available.
Yes, the quirk is in the protocol, not your code.
I would rather not reveal exactly what I found so that the advantage will not be nullified by multiple teams jumping on the same bandwagon.:o :ahh:
Mike,
My concern with posting what I did was that I would be detracting from the work you had done. By no means did I intend to minify your solution. The software only solution Kevin provided is just that, another solution.
Between these two options, teams should be fairly well covered.
This is a great idea, our team fried the IR board, and we have not looked for a replacement because of cost, but we have an ability (PCB knowledge) to make the circuit board through http://4pcb.com with their Student Discount fee.
A second and less obvious diode is common across the Vin and Vout pins of the regulator. This is a common design practice (but interesting left off the app note for the LM79L05) This diode protects from transients during power down. Imagine the device is powered up and then Vpwr is suddenly (0V) removed. This second protection diode is required to discharge the filter caps on the output by safely shunting the current around the regulator. Without this diode, the Vreg will be stressed with reversed polarity. VReg are hyper sensitive to any reverse voltage. As engineers we need to think beyond the “run” mode and also consider the power up and power down modes.
---<|-----
Mode Vin Vout Diode
Run 12V 5V Reverse Biased
Pwr Up 0->12V 0->5V Reverse Biased
Pwr Down 0V 5->0V Forward Biased
Hope this was helpful. Sorry the feedback is “late”
Excellent point, that is one reason why the 33uF cap is on the upstream side of the reg, that and to also filter out ripple.
That is a great suggestion. The reason I stuck with the two row config is so teams that have already made custom wiring would not have to make a new harness. If they need a replacement they can just swap out boards.
Same reason as above.
This is a great suggestion. I will see if I can fit a T-220. We bypassed the reg all-together and just run off of the 5 volt dig outputs +5v rail. The rc has a 5 volt reg built in. Why not use it. If you do, you have the advantage of not dealing with transient V-drop from overloading the 12 volt battery. Any time Vbatt drops below Vdropout the backup battery kicks in. You have near perfect regulation. Also, you can wire the board directly to the rc without a breaker as allowed by the rules. This saves weight and I know how precious that can be to those mechanical people.
Mike.
I agree that a diode across the regulator with the cathode at V in is appropriate with a larger output capacitor. Not sure if that needs to be the case here. However, I would recommend a 10 mfd near the chip just to protect against transients when the LEDs turn on. I like the use of the RC 5 volt rail but the you should then keep the big cap in place to filter the power. One team at Boilermaker had a short on the board pull down the RC 5 volts. Perhaps a small fuse is in order as well.
Just before the WI regional I ran across an article in SERVO magazine about a spinning battle bot that needed to get IR commands from its operator. He took a 18" x 1" lexan rod and drilled and tapped a hole thru the center. He then fitted the rod over the IR sensor. IR form the remote would be collected by the rod and the threads acted as prisms to direct the light toward the sensor. He got 360 degree coverage and increased the light capturing area from 1/4 square inch to 18 square inches.
We used this on our bot and doubled the range of our IR board.
BC
Team 1675