New problems keep coming up and its not working. Help!

I have the trackmate hardware which works fine with PC lapcounter. I put optical sensors for pit in and pit out in the Scalextric track so the cars will break the light sensors in these guys and it is hooked to a phidget board. I am running Windows XP Pro, latest. I have the latest PC Lapcounter software too.

Problems I am having is that the cars sometimes activate the sensors, sometimes don't. It isn't consistent. If we go fast through them, the sensors won't activate mostly, so we are going slow into them, and sometimes it works. The pitout sensors are doing the same thing.

Has anyone made this setup work using optical sensors in the track? I sure would like to know how you did it. Or is the lightbridge the only reliable way to go?

Thanks for helping. This is getting really frustrating and cannot be used in a race at this point.

Racer Randy
www.beavertonslotcarclub.com

Hi Racer Randy,

We have had the same problems with the optical sensors: they react too slow and most of the times fast passing cars do not activate the sensor... at least that is what it looks like...
But in fact is it not the optical sensor but the Phidget card that is the bottleneck.

Problem is this: the "pulse" that the optical sensor gives to the phidget card is simply not long enough. The Phidget card only reads every 0.125 secs if there is any pulse... but the optical sensors are much much much faster... so when the Phidget card tries to read something : the "pulse" from the optical sensor is already 'gone'.

Because you can not change the Phidget card: the solution we made is a small hardware adaptor to make sure the "pulse" which comes from the optical sensors is long enough. This is done with the NE555 (timer IC).
Because we use the Phidget 8/8/8 i.s.o. the Phidget 0/16/16 we also had to adapt the pulse from positive flank to negative flank triggering.

Please have a look at our previous postings about the 8/8/8 electric schema

So this also means that other types of sensors do not solve your problem !

Succes !

Danny
http://www.slotmanege.nl (sorry, only in Dutch)

Randy,

I forgot to mention that the color of the guide-flag of your slotcars also is of importance !!
Make sure you use BLACK colored guide-flags.
Our used sensors do not register blue or white guide-flags.

Regards,
Danny

Hi Danny,
Thanks this is helpful. You mention that the white or blue guide flags don't work as well as the black ones. Is this because the black ones are deeper or is it the color itself. If the color, have you tried painting the guide flags black?

Other than that, is everything working really well now? thanks
Racer Randy

Hi Danny,

You are using it for lap counting and seem to need only 1 sensor per lane. We are using it for Fuel Management and need a pit out sensor too. What are your thoughts about doing a pitout sensor too? Would it be the same schema?

Racer Randy

Danny - You wrote the following:
<snip>

slotmanege wrote:

Hi Racer Randy,
The Phidget card only reads every 0.125 secs if there is any pulse... but the optical sensors are much much much faster... so when the Phidget card tries to read something : the "pulse" from the optical sensor is already 'gone'.

<snip>

I believe you meant to state that the Phidget card reads a pulse every 1/125 sec, which would be .008 sec (not .125 sec)
Just wanted to set the record straight.
Foiler

@Foiler : Sorry, you are right... my mistake. What I ment is indeed :

Quote :

that the Phidget card reads a pulse every 1/125 sec, which would be .008 sec (not .125 sec)

@Randy : About the guide flags. It is color related and not related to the heights of the guideflags because the heights are the same.
When painting the blue of white guideflags to black it works fine... but best is to take a black one because the paint can wear off...

Yes, we have only one sensor per lane and we only use the optical sensors for lapcounting on our clubtrack. We do not use pitstops or fuel management.
We have had some problems with EMC (electro magnetic influences) which resulted in counting laps even if there was no car passing the sensor... These problems were solved by using good shielded data-cables and moved the phidget card further away from our power units.
Now all works just fine ! And we have 100% accuracy
Last saturday there was a big race at our track called "RacebaanForum race". Dutch and Belgium drivers raced in a teamrace for several hours.
Even then the system did work as it should, no errors, no failures !

I do not know how others use the pitstop and what sensors are used there? I think one sensor for pistop is enough? You just want to register the car if it is in the pit? Why do you want to detect the car leaving?
If you do want to make use of two sensors for pit-in and pit-out : yeah the same diagram could be used, no problem. It should work !

slotmanege wrote:

I do not know how others use the pitstop and what sensors are used there? I think one sensor for pistop is enough? You just want to register the car if it is in the pit? Why do you want to detect the car leaving?
If you do want to make use of two sensors for pit-in and pit-out : yeah the same diagram could be used, no problem. It should work !

Here is the setup...
- The track is using deadstrips with Trackmate for lap sensing/timing.
- Randy's preference is to avoid light bridges.
- We have pit-in sensors, installed in the slot, but to actualy stop so the the guide flag is blocking the them is near impossible.
- The desire is to allow partial fuel by interupting the pit stop.
- We can either use a latching circuit to hold the pit-in signal and let PCLC to generate the pit-out, after a driver uses a switch to unlatch the circuit
OR-
-We use a pit-out sensor to detect a car leaving, triggering PCLC to cease fueling.

We have tried the switched latching circuit and have encountered problems with adjacent lanes unlatching upon passing the pits.

In trying the Pit-in/Pit-out sensor aproach, the .008 sec deadband on the phiget is creating issues.

I will have to use a NAND gate in front of your timer circuit so that the "Pull-Up" signal generated by my sensors are instead converted to a pull down signal, triggering the NE555/558 and the subsequent NAND gate and thus the phiget

I could do a little salvage and resoldering, but I perfer to design something I can plug in-line to the installed setup.

Glenn