Transponder Design
#332
Tech Apprentice
Was googling a bit and found the specs on the RC4 transponder. Some of this is interesting to compare with the Cano transponder.
Code:
Specifications Dimensions (WxLxH) : 16x19x6 mm (approx. 0.7”x0.8”x 0.2”) Weight : 4 g. Weight incl. Holder : 4.7g. Humidity : max. 90% relative Max. speed : 120 km/h / 75 mph. Temperature range : 0-50 c / 32-122 F Signal transfer : magnetic induction Code resolution : 1.5 ms Transponder position : max. height 15 cm / 6’’ Power : Feed from free receiver channel (ch3 or batt) Connection Plug : connect using a standard Servo plug Power consumption : 19mA @ 6VDc Operating voltage : 2.8 – 16 VDC Wire insulation : Aerospace/MIL-spec high-temp PTFE (Gasoline/Nitro resistant)
#333
Tech Rookie
Hi all,
I found out about this thread about a week ago and absolutely loved it. I took everyones efforts and put it in a new design to what I think (from my background) will work best.
I followed Howard's approach on creating a 5MHz carrier because in my opinion it's not worth the effort to put this burden on an MCU for a simple binary modulation as this one. The way how Howard designed the spiral loop antenna for this application also receives my preference in terms of output strength (when tuned properly).
On our track we only drive 2S up to 6S powered cars so I opted for a power management solution that allows receiving power from any of these battery packs. It is very much possible to design a tiny power management circuit for a 1S pack as well but in my approach where I have very limited board space I had to make a choice because increasing the input voltage range from 1S to 6S requires more board space. You can actually keep my current design and simply swap U3 with an ultra low dropout voltage regulator that has the same pinout and SOT-23 package.
I like the idea how firmware in ESC's can be configured/upgraded through a 1-wire interface. I was thinking about doing something similar by first flashing the MCU with a bootloader that accepts new firmware to be uploaded using 1-wire protocol. This way you don't need to have physical access to the PCBA for reflashing the MCU but is everything reprogrammable through the 3-wire cable that is brought out. Therefore I went with an ATMEL because I have most experience with this manufacturer for applications such as this one.
I shrunk everything down to a 20x20mm SMT only PCBA. This is still hand solderable but I admit that it is quite a challenge.
I'll build a couple units early August. Send me a PM if you want to join in with the programming. After tuning the first units I'll send one out to those that believe they can be of any help (You will need an AVR ISP capable flasher to flash the MCU directly and an Arduino nano for example if you also want to test/use the 1-wire upload ). I'm sending out 5-8 units.
I use a handy tool to flash the MCU directly through the ISP interface because then I don't have to bring out any additional programming pads or a connector. There is simply no room for it anyways. Since I'm new here and this is my first post I am not allowed yet to post URL's so if you google "ATmega firmware flashing tool" you can see what I am talking about.
Attached are the PCB layers. I'll post a 3D render and schematics in the next post.
Any comments/concerns/remarks are welcome.
I found out about this thread about a week ago and absolutely loved it. I took everyones efforts and put it in a new design to what I think (from my background) will work best.
I followed Howard's approach on creating a 5MHz carrier because in my opinion it's not worth the effort to put this burden on an MCU for a simple binary modulation as this one. The way how Howard designed the spiral loop antenna for this application also receives my preference in terms of output strength (when tuned properly).
On our track we only drive 2S up to 6S powered cars so I opted for a power management solution that allows receiving power from any of these battery packs. It is very much possible to design a tiny power management circuit for a 1S pack as well but in my approach where I have very limited board space I had to make a choice because increasing the input voltage range from 1S to 6S requires more board space. You can actually keep my current design and simply swap U3 with an ultra low dropout voltage regulator that has the same pinout and SOT-23 package.
I like the idea how firmware in ESC's can be configured/upgraded through a 1-wire interface. I was thinking about doing something similar by first flashing the MCU with a bootloader that accepts new firmware to be uploaded using 1-wire protocol. This way you don't need to have physical access to the PCBA for reflashing the MCU but is everything reprogrammable through the 3-wire cable that is brought out. Therefore I went with an ATMEL because I have most experience with this manufacturer for applications such as this one.
I shrunk everything down to a 20x20mm SMT only PCBA. This is still hand solderable but I admit that it is quite a challenge.
I'll build a couple units early August. Send me a PM if you want to join in with the programming. After tuning the first units I'll send one out to those that believe they can be of any help (You will need an AVR ISP capable flasher to flash the MCU directly and an Arduino nano for example if you also want to test/use the 1-wire upload ). I'm sending out 5-8 units.
I use a handy tool to flash the MCU directly through the ISP interface because then I don't have to bring out any additional programming pads or a connector. There is simply no room for it anyways. Since I'm new here and this is my first post I am not allowed yet to post URL's so if you google "ATmega firmware flashing tool" you can see what I am talking about.
Attached are the PCB layers. I'll post a 3D render and schematics in the next post.
Any comments/concerns/remarks are welcome.
Last edited by wova21; 07-25-2016 at 06:50 AM.
#334
Tech Rookie
Schematic
Last edited by wova21; 07-25-2016 at 01:12 AM.
#336
Tech Rookie
PCB's are ordered. I started to work on a 1-wire bootloader for the ATmega when I ran into this-> blhelisuite.wordpress dot com
These guys have already figured it out for multirotor ESC's and since I'm using the same MCU it looks like I can just go ahead and flash their bootloader to enable the 1-wire programming interface. They provide the source which is awesome. I had to increase the PCB from 20x20mm to 20.3x20.3mm to allow panelizing the boards. I got in trouble with the edge clearance. I use standard pool design rules offered by most of the PCB manufacturers around the world which means that they are very affordable to manufacture.
total BOM cost when ordering small QTY's (PCB+parts+servo cable) is about 10USD.
These guys have already figured it out for multirotor ESC's and since I'm using the same MCU it looks like I can just go ahead and flash their bootloader to enable the 1-wire programming interface. They provide the source which is awesome. I had to increase the PCB from 20x20mm to 20.3x20.3mm to allow panelizing the boards. I got in trouble with the edge clearance. I use standard pool design rules offered by most of the PCB manufacturers around the world which means that they are very affordable to manufacture.
total BOM cost when ordering small QTY's (PCB+parts+servo cable) is about 10USD.
Last edited by wova21; 07-25-2016 at 01:14 PM.
#337
Tech Rookie
First proto's are in.
I got the first one finished and powered up. I still have to tune the tank but here are the pics of what it looks like.
On the software side I am able to read in ID values stored in EEPROM so it's easy to store a new ID value in the EEPROM through the one wire interface. This is still under development but so far good progress has been made.
I know the thread is rather old and not that active anymore but let's see if we can get some of you excited again to get involved again
On the software side I am able to read in ID values stored in EEPROM so it's easy to store a new ID value in the EEPROM through the one wire interface. This is still under development but so far good progress has been made.
I know the thread is rather old and not that active anymore but let's see if we can get some of you excited again to get involved again
#338
I got the first one finished and powered up. I still have to tune the tank but here are the pics of what it looks like.
On the software side I am able to read in ID values stored in EEPROM so it's easy to store a new ID value in the EEPROM through the one wire interface. This is still under development but so far good progress has been made.
I know the thread is rather old and not that active anymore but let's see if we can get some of you excited again to get involved again
On the software side I am able to read in ID values stored in EEPROM so it's easy to store a new ID value in the EEPROM through the one wire interface. This is still under development but so far good progress has been made.
I know the thread is rather old and not that active anymore but let's see if we can get some of you excited again to get involved again
#339
Tech Apprentice
Very nice progress indeed. I invested a bit of effort into tweaking Howard's design for my needs, i.e. compatibility with higher voltages and I have three copies of it.
However the drawback of Howard's design is its size and weight. So I might be interested in getting some of yours as well if that is feasable. Did you solve the issue of getting the transponder compatible with the RC4 decoder?
However the drawback of Howard's design is its size and weight. So I might be interested in getting some of yours as well if that is feasable. Did you solve the issue of getting the transponder compatible with the RC4 decoder?
#340
Tech Rookie
Very nice progress indeed. I invested a bit of effort into tweaking Howard's design for my needs, i.e. compatibility with higher voltages and I have three copies of it.
However the drawback of Howard's design is its size and weight. So I might be interested in getting some of yours as well if that is feasable. Did you solve the issue of getting the transponder compatible with the RC4 decoder?
However the drawback of Howard's design is its size and weight. So I might be interested in getting some of yours as well if that is feasable. Did you solve the issue of getting the transponder compatible with the RC4 decoder?
Therefore I removed the xtal and replaced it with a resonator which is much much smaller.
I went with a 10MHz resonator and generate the 5MHz from a timer output (prescaler at 2) giving me more clock cycles between each transmission "bit" and by using timer output the machine cycles are still in sync with the carrier wave.
Secondly I was comparing my output strength with a Mylaps Hybrid PT and I noticed that my output was about 3dB less. In the new PCB I gave in trace length for trace width and went with 2oz of copper instead of 1oz resulting in a much lower series resistance for my inductor spiral antenna which in turn gives me a much higher output.
The overall thickness at the moment with eliminating the xtal is about 3.6mm. (I went with a 1mm PCB not a 1.6mm)
I have not tested RC4 compatibility. I'm merely focussing at the moment on having a user friendly version which doesn't require a complex IDE or isp/jtag pods to load in a new number. I'm currently stuck at the bootloader which accepts new parameters to be programmed into eeprom over a 1 wire interface. If can store them directly in eeprom and read them from there but it still requires an AVR programmer to push them into eeprom.
I actually live about a 1hour drive from the MYlaps HQ but I don't think they would be of much help. I haven't thoroughly analyzed the transponder output of an RC4 PT so I can't give you any feedback on that.
Anyways. I had to put in a new order at work and there was some room left on the PCB panels so I filled them up with the new PCB so I have about 150pcs laying around. So let me know if you want a couple or if you are interested in the gerbers. I'll upload the new schematic later as well.
Cheers.
#341
Tech Rookie
Revised PCBA
Hi,
I attached the updated schematic. Don't be thrown off with the amount of caps present at the LC tank. I haven't actually mounted them all. I have been playing around with a low pass filter on the input but I have not found any significant improvements over Howard's main design. I'll take and upload some pictures of this version built. I currently leave them at the office because otherwise I would get into it every night. It's very thin and the output is great compared to my first build.
My version accepts up to 6S Li-Po's and I have stressed it up to +30VDC for a decent amount of time which does not seem to be a problem. It's also within the specs of the part so that was expected. I also added that P-channel MOSFET because we all have inverted the polarity of our battery packs at least once in our lives. And if blowing up an ESC isn't costly enough already let's try to protect what's left in the circuit .
Let me know what are you looking for Barsk and I can probably help you out. Gerbers, bare board, built board...
I attached the updated schematic. Don't be thrown off with the amount of caps present at the LC tank. I haven't actually mounted them all. I have been playing around with a low pass filter on the input but I have not found any significant improvements over Howard's main design. I'll take and upload some pictures of this version built. I currently leave them at the office because otherwise I would get into it every night. It's very thin and the output is great compared to my first build.
Let me know what are you looking for Barsk and I can probably help you out. Gerbers, bare board, built board...
#342
Tech Rookie
Registered on this forum just to reply to this thread. Super interested in all that is happening in here and look forward to as much info as wova21 and howardcano can provide. Thank you for all of the effort already.
I help run a volunteer motorcycle race league and the AMB pricing is killing us since we already charge racers literally the lowest cost we can because we all just love to race (which is less than 1/2 the cost of all other leagues here in california). Some of our xponders are starting to go bad (more than just batteries) and replacements will be needed in the future.
I help run a volunteer motorcycle race league and the AMB pricing is killing us since we already charge racers literally the lowest cost we can because we all just love to race (which is less than 1/2 the cost of all other leagues here in california). Some of our xponders are starting to go bad (more than just batteries) and replacements will be needed in the future.
#344
I got the first one finished and powered up. I still have to tune the tank but here are the pics of what it looks like.
On the software side I am able to read in ID values stored in EEPROM so it's easy to store a new ID value in the EEPROM through the one wire interface. This is still under development but so far good progress has been made.
I know the thread is rather old and not that active anymore but let's see if we can get some of you excited again to get involved again
On the software side I am able to read in ID values stored in EEPROM so it's easy to store a new ID value in the EEPROM through the one wire interface. This is still under development but so far good progress has been made.
I know the thread is rather old and not that active anymore but let's see if we can get some of you excited again to get involved again
#345
Tech Initiate
iTrader: (2)
Hi,
I attached the updated schematic. Don't be thrown off with the amount of caps present at the LC tank. I haven't actually mounted them all. I have been playing around with a low pass filter on the input but I have not found any significant improvements over Howard's main design. I'll take and upload some pictures of this version built. I currently leave them at the office because otherwise I would get into it every night. It's very thin and the output is great compared to my first build.
My version accepts up to 6S Li-Po's and I have stressed it up to +30VDC for a decent amount of time which does not seem to be a problem. It's also within the specs of the part so that was expected. I also added that P-channel MOSFET because we all have inverted the polarity of our battery packs at least once in our lives. And if blowing up an ESC isn't costly enough already let's try to protect what's left in the circuit .
Let me know what are you looking for Barsk and I can probably help you out. Gerbers, bare board, built board...
I attached the updated schematic. Don't be thrown off with the amount of caps present at the LC tank. I haven't actually mounted them all. I have been playing around with a low pass filter on the input but I have not found any significant improvements over Howard's main design. I'll take and upload some pictures of this version built. I currently leave them at the office because otherwise I would get into it every night. It's very thin and the output is great compared to my first build.
My version accepts up to 6S Li-Po's and I have stressed it up to +30VDC for a decent amount of time which does not seem to be a problem. It's also within the specs of the part so that was expected. I also added that P-channel MOSFET because we all have inverted the polarity of our battery packs at least once in our lives. And if blowing up an ESC isn't costly enough already let's try to protect what's left in the circuit .
Let me know what are you looking for Barsk and I can probably help you out. Gerbers, bare board, built board...