Since we all seem to be in a sharing spirit, let me make a contribution:

For those interested in creating a transponder from my designs in this thread, I will make available software for the PIC12F683, for a single ID number. You can program the PIC directly from the hex file provided (if you have a programmer!). I will also include the source code for those of you who are curious.

Send me your email address if you would like to receive the zip file containing this software. You may use everything I provide for your own personal, non-commercial applications.

To many components. and i have lots of atmel cpu that I'm used to program, never used pic.

MRT transponder has numberd that does not need a status message. Works great with mylaps with last firmware.

For example:
5843805
E29BAD5D7B4479877F0300

8087916
0DAEADB24989B964B33340

5781589
E274A37298704A488CF340

I also noticed when testing random numbers i found that 8087916 have no ghost number on my old decoder, don't know about the other numbers. I thought this might have something to do with the unknown bits in the data because they are mostly 1 in this number while 0 in other, but encoding this number with 0 as unknown data also works without ghost. Does this number look like a valid status message just by random luck?

I still haven't found a single status message in my real transponder ( ambrc dp model) only hundreds of id message so i have no idea yet how they look.

it looks that new decoders checking also 'ghost bits'. Older decoder probably don't care.
In my encoding sw you can specify 'ghost bits' can anybody here test this number with several ghost bits combinations with new decoder?

Here is my transponder design so far with big components, built on height due to the fact that the cpu in holder will get high anyway.



Antenna design is terrible at the moment and the signal is weak but I'm working on it. Caps and number of turns will change.

This is a version with input voltage capable of 5-12v

This thread has reached a good milestone: 30,000 views! That's about half of what the decoder thread has, and is a view-to-post ratio of about 150. So here's a big "THANK YOU" to everyone for your interest, and for sharing your ideas and experiments. I hope you've had fun and learned a few things here. I certainly have!

Hello Howardcano,
I am very impressed by your realisation ! Very interesting.
Congratulation ! and as you say, i have fun and i learn few things.
Can you please send me the PIC program with commentary.
And above all, can you explain me how did you calculate L1, CT1, CT2, R4 and R5 please ? Electronic part interest me a lot.
(Sorry for my poor english, i am french.. )
I thank you very much, and for all other people for this project.
David

Hi David,

Welcome to the thread and to RCTech!

Send me a private message with your email address and I'll send a zip file to you. You can send a private message by clicking on "Trader Rating" under my RCTech username, and then selecting "Send a private message to howardcano".

R4 and R5 were selected to restrict the output current of the 74AC86 gates to a safe value.

The number of turns for L1 was determined by the maximum that I could fit on the double-sided circuit board while still having room for the components.

CT1 and CT2 were chosen so their series combination would resonate with L1 at the transponder's carrier frequency of 5MHz.

The ratio of CT1 and CT2 was then chosen to set the desired Q of the circuit (which is a compromise between harmonic attenuation and bandwidth necessary to follow the phase modulation).

It took a couple of iterations to arrive at the capacitance values indicated, since both the value and Q of the inductor are affected by the components within its aperture.

A few people have PM me about transponder designs and the missing values of my design test, my latest version that works ok and uses minimal complexity looks like this:


The L1 is just to visaly represent a wire don't use that part. I soldered a solderpath i turn around the pcb and then connect antenna wire so it have something on board also if wire break.
The big 1:5 cars at our track have the radio box at least 10cm off the ground and have a big aluminium chassis that blocks radio very well, so this version with a wire sticking out as the antenna worked extremely well. But i guess the radio signal is very dirty and the q&i that i still don't know what they are are probably wrong also. But it works that is all that matter to some.
And parts and cost is minimal. I don't have 1000 different cap values to test the resonating antenna loop designs and i always loose count of the turns so this worked better for me
And it don't have the DC-short problem that i have to solve with resistors on the output before.

I will record radio spectrum as soon as i get time to compare with other design.
But with a normal radio scanner, i can hear this transponder 1m from the car and other antenna is hearable 5-20cm from car. (decoder loop don't hear it 1m off the car)

EDIT: I did the picture in online software as fast as i could before time demo ran out, will draw in proper software another time.

Thank you for this. I'm hoping to have a go building some of these once I've finished my current experiment (charger custom firmware).

I've got some basic questions if that is ok.

Your schematic shows the attiny24 chip, the original had the attiny25. Will both chips work?

Regarding programming. I've just successfully programmed some custom firmware to a charger (https://github.com/stawel/cheali-charger) using a USB ASP adapter. Do you know if I can use the same adapter to program the attiny24?

It would also be great if you could show us a picture of your transponder. I'm interested to see the free hanging wire.

Thanks again,

Tony.

Typo, is attiny25

I've built one and am hoping to try it this evening.

A quick question on fuses. I tried setting L fuse to ff, but it seems to have bricked a chip (I get no reply from the chip when trying to identify it using avrdudess). Am I doing something wrong?

Tony.

If you set L fuses to FF the chip MUST have a crystal or oscilator to be reprogramed. And the bad part is that the 20mhz is to fast for most programmers, so use a 8mhz.

My first attempt didn't work (the decoder did not pick up any signal from my transponder). But I suspect it's because I haven't programmed the chip properly (I used a USBAsp programmer).