Help for restoring an old RC Lap Counter

[marcoeffe]

Hi everyone, I'm new in this forum and this is my first post. I'm a software developer but electronics has always been one of my favourite hobbies. I don't have experience in RF circuit design, and for this reason I'm here trying to find someone that can help me in this project. I'll try to be coincise as much as possible.

I have an old RC Lap Timing system which has been built by someone 20+ years ago. This transponder system is composed by the usual small plastic cubes which are installed onboard (on rc cars) and are recognized by the receiver station when the rc car passes over the antenna loop (a wire which is layed on the track). These cubes are died because of the leaky battery that after all of these years has ruined all the internal circuits, but the receiver still works pretty well.

What I'm trying to do is to rebuild the transponder cubes using an alternative solution, maybe using modern oscillators, or in the worst case reusing the existing crystals. It's hard to figure out which are the components due to a glue which covers everything, but after taking a look to the circuitry of transponders and receiver station, I can say the following:

- The transponders are transmitting just a plain signal on the crystal frequency, or multiples, since there are not particular components apart from a JFET transitor and an inductance. Crystals are in the range of 5,3 to 5,6 MHz.
- The receiver is just an array of narrow filters tuned on the frequency of each transponder cube, so no decoders or particular things. This is compatible with my hypothesis that the transmitted signal is just a sine wave or at most a square wave.
- I tried to generate an signal using an RF generator but I didn't manage to trigger the receiver, even in sweep mode, I didn't get any "transponder" pass. The generator I have is capable just to generate signals from 35 MHz, so I'm assuming that the used frequencies are below this value, even if multiple of the crystal base frequency.

I'd like to build something that can completely replace the existing cubes, and after some research I saw that there are programmable oscillators or the easiest adjustable ones like the LTC1799 which I think could fit my needs. Now, even if these transponders are working in a very close range (max 60cm) I guess that an amplification stage is still needed. This is the only missing part for me, so I'd like to ask you guys some help in this way.

Obviously if you have any comment, question or remark, feel free to reply!

Thanks in advance!

 

[augustinetez]

The LTC1799 is a bad choice - it is not stable and drifts all over the place (frequency wise).

The original cubes used crystals for a reason - stability - which is needed for a narrow passband receive system.

Next question is how big are the cubes - that will point to what might be able to be used as a replacement.

Crystals in anything but standard frequencies are expensive nowadays, so you will likely be looking at things like clock generator chips or similar - eg Si5351, Si570 and the like, but they do require a microcontroller to run them unless you order the preprogrammed version (costly) - this is but one example of using the above SI5351 https://www.vk5tm.com/starlo.php.

One other minor problem at the moment is that a lot of clock chips ( and many others) are in very short supply with availability not until late 2022.

In answer to the need for some amplification of the signal - probably.

 

[marcoeffe]

The LTC1799 is a bad choice - it is not stable and drifts all over the place (frequency wise).

The original cubes used crystals for a reason - stability - which is needed for a narrow passband receive system.

Next question is how big are the cubes - that will point to what might be able to be used as a replacement.

Crystals in anything but standard frequencies are expensive nowadays, so you will likely be looking at things like clock generator chips or similar - eg Si5351, Si570 and the like, but they do require a microcontroller to run them unless you order the preprogrammed version (costly) - this is but one example of using the above SI5351 https://www.vk5tm.com/starlo.php.

One other minor problem at the moment is that a lot of clock chips ( and many others) are in very short supply with availability not until late 2022.

In answer to the need for some amplification of the signal - probably.

Hi, thanks for your response and the explanations.

The transponders are very small, 4x3 cm, but I can easily reprint a different enclosure if needed.

At this point I'd say that the best solution would be reusing the existing crystals for the reasons you mentioned. I found some schematics online, but obviously the supply voltage is always at least 5V, while the battery that they had is a 1/2AA NiCd, so 1.2V. I was exploring the possibility to use a different battery, but I'd like to keep the same ones so I can reuse also the charger.

This is one of the schematics I found:

Not sure that it can fits 5MHz crystals, apart from the voltages that would be different for sure.

Thanks again.

 

[augustinetez]

That should run the 5MHz crystals OK.

If you are going to refurb this, I would take the opportunity to update it and use 3.7V LiPo packs.

You can get small flat ones with enough capacity to run for a day in this application that would easily fit in the space.

It would then only be a case of adjusting the resistor values slightly to make your circuit above work reliably at the lower voltage or there are op-amp circuits that will do the job.

Here is one example from Analog

 

[marcoeffe]

That should run the 5MHz crystals OK.

If you are going to refurb this, I would take the opportunity to update it and use 3.7V LiPo packs.

You can get small flat ones with enough capacity to run for a day in this application that would easily fit in the space.

It would then only be a case of adjusting the resistor values slightly to make your circuit above work reliably at the lower voltage or there are op-amp circuits that will do the job.

Here is one example from Analog
View attachment 133684

Oh, thanks a lot!

I'll try to simulate in LtSpice the transistor solution since it would be the cheapest way.

One last question, do you think these solutions require an additional amp stage or I can connect an antenna directly on the output?

Thanks.

 

[augustinetez]

Is there any chance of tracing the circuit of an original?

That should give you an idea of what power output might have been - at a 1.2V battery supply, it wouldn't have been much and you also don't want too much signal otherwise it will trip the detector before it reaches the timing/finish line.

I would start by connecting the antenna direct and see if it works as it should and proceed from there.

 

[marcoeffe]

Tracing the original circuit is quite difficult. Everything is stuck in a whole block of glue and if I try to apply some hot air, everything falls apart.

I'll try to connect an antenna directly, hopefully I can compensate the signal using the RF Gain control available on the receiver.

Thanks!

 

[augustinetez]

That's unfortunate (thanks for trying).

 

[marcoeffe]

Quick update of the status of my project. I'm getting very close to a solution, and actually everything seems to work perfectly. I found a super simple schematic on this page which is a CW transmitter capable to cover several miles.

This is the schematic:

The author wrote that he managed to generate a decent signal even with 3v, so I tried to create a quick prototype:

And it works!

Transponder #3 is recognized!

I tried with 3.6v and seems to be fine, even if the power definitely drops a lot but is still recognized in a very close range.

Now my improvement would be to try to make it work with a single 1.2 cell. I know that there are LiPos that I can use but I'd prefer to keep the existing charger since the project budget is quite small. Any suggestion on how to achieve this final goal?

Thanks in advance!

 

[augustinetez]

Starting point:-

Drop the 100k resistor to 47k

Drop the 1k resistor to 330 ohms - this may need dropping further but I wouldn't go below 100 ohms

It will be a juggling act between the two resistors as to where the circuit runs reliably.

 

[unclejed613]

Everything is stuck in a whole block of glue and if I try to apply some hot air, everything falls apart.

spray the block of hot glue with freeze mist until it's completely frozen, and the hot glue gets brittle and crumbly and is easier to remove