THSCC Forums

Thanks for the input everyone - Keep the ideas coming!

One other factor that needs to be considered is sampling rate and the accuracy of a trigger.

The native system has 0.001 second accuracy. If the "black box" we're considering samples at any longer interval, we're screwed.

Is there a sampling frequency in the RS-232 standard?

As for price - we don't need to cheap out. I would balk at anything more than $500 or so right now.

_________________
Mike Whitney
whit32@gmail.com, 919-454-5445
V10, V8, V8t, I6, I6, V6, F4t, I4, I4, I4, I4, I2, 1, 1

Wouldn't most (all?) of these wireless solutions introduce a latency into the timing picture? Obviously at first glance the transmission speed is the same for everybody but in order for something like this to function in the harsh outdoors, don't most (all?) of these things have a built in error correction protocol?

If the receiver misses a transmission (or gets one that has been corrupted), most of the transmitters are set up to retransmit which would introduce a timing error into the system. In a sport that milliseconds count, I don't see this working too well.

If a retransmission could be detected by the timing software, perhaps a rerun could be issued as a possible solution. But is this just introducing more complexity into the problem along. Hopefully this would be kept to a minimum but with some of the sites we run at (like the active airports), this could be a potential big problem.

Just something else to think about anyway...

I can think of a few potential solutions but I need to do some digging around to see if they are feasible and cost effective.

Anybody care to give a detailed and accurate picture of the current timing layout? Including the software layout?

Jim

Are we inclined to do some sort of joint development venture (meaning shared costs) with JAC's? Seems like the club has a bunch of available credible talent to assist the JAC's guys.

_________________
Dick Rasmussen

FS 50 2018 Mustang GT

We need to stay with a very simple system - no error correction, no retransmission, no protocol conversion. Just an antenna that stays quiet until +5v is detected, an RF pulse is generated, picked up on the other end by the receiver, which regenerates the +5v signal. If the sampling rate of the transmitter is high enough, latency will be the same as long as we don't move anything!



Let's not confuse the goal here. No software changes. Just tx and rx boxes which can replace the wire and will "fool" the JAC box. Assume it's just a simple +5v square wave pulse for now.

I think there's got to be a pre-packaged system out there which will do this. It's just a matter of finding it or finding someone who knows [/quote]

_________________
Mike Whitney
whit32@gmail.com, 919-454-5445
V10, V8, V8t, I6, I6, V6, F4t, I4, I4, I4, I4, I2, 1, 1

So we just let the timing people deal with the inevitable transmission glitches that come up? Any idea how often that happens with the current system?



Actually, I had my own reasons for asking this question that didn't really pertain to this thread. Sorry for the partial hijack...

Jim

RS232 is just a voltage level. *If I remember correctly* 5v becomes -12 volts, and 0 becomes 12volts. It is all analog. (the 12v would be created by a maxim(or whatever) 232 chip). It is just a voltage level standard used to transmit data over long distances (max 250ft?) of cable. The higher voltage diff(24v) is used because it is better than just 5 volt difference. So whatever the sensor would feed it (0 or 5) it converts to -12 or 12. So your sampling question would all be based on the wireless unit used, not the black box.

As far as error checking, if there isn't any you only have the time the signal is "open"(car going across the line) to convert to RS232 voltage, then transmit, recieve, and then be picked up by the unit in the bus.

_________________
2001 Honda S2000 - SOLD
2012 Boss 302
2003 BMW 330i

Actually I believe RS232 is defined as a protocol used to transmit digital data over an analog medium. While the voltage level is reasonably important, it's actually defined in a fairly broad to leave a lot of room for noise and errors (or lack thereof). I'm pretty sure the actual range is +- 15v on the transmitter side and +- 25v on the receiver side.

A lot of different devices use different voltage levels for the actual transmitting and receiving leading to the occasional incompatibility or glitch between them even though they are both within "spec." Ever have a laptop not communicate with desktop over a serial line even though all the settings are the same on either end? Generally it's the voltage levels that are causing the grief and there isn't much you can do about it.

The actual transmission is defined as a start bit, some amount of data bits, a parity bit if needed/wanted and some number of stop bits. To actually transmit, both sides have to agree on a all this stuff as well as a speed; otherwise, strange and generally undefined things happen. The speed of the transmission is probably irrelevant to this application because of the low amount of data needed.

Damn, I'm a nerd! This is all from a fairly hazy memory from my long-ago EE days so please correct me if I've goofed anywhere. Anyway, most of this is irrelevant to the discussion so I'll just leave it at that.

If all you want is a simple circuit that will transmit a signal to drop a line low for an instant (to trigger the timer), I could probably whip one up fairly easily and cheaply. I was just trying to be clear on what you were asking for and my engineer side wanted to over-complicate the situation as usual.

While the 1 mile range might be a problem, there are a couple of ways around that without building/using a high-powered transmitter. Would it be sufficient to have a shorter wireless range but be able to add to it with a cable? If you had a few hundred feet of cable on either end of the transmit and receive side, you could seriously extend the range using low-powered wireless stuff. Stuff being a technical term! Seriously, it would help solve the range problem and keep the hard and heavy stuff (like a car battery) away from the timing lights.

Another possible wired solution if the current wires are just to short would be to use a signal conditioner or something to extend the range of the current equipment. Also, if you wanted to use the RS232 approach but stay wired for greater reliability, I believe the range on RS232 can approach 3000ft if used with a low baud rate (in the 2400 range). With 2 transmitter/receivers in series, you could easily cover a mile.

I'm sure we could come up with couple of other options depending on what you're really after. Let me know if you want me to further pursue any of this and I'd be happy to.

Jim

One possible place to look is what do dragstrips use? I don't think all of them are hardwired, and they have to have precision timers.
Also maybe contact the land speed record people that use Laurenburg to see what they use, they must have long range precision, as well as portability and realistic costs.

_________________
SPIN or WIN!
there's no glory for going slow.

This got me thinking about another idea...it breaks the boundaries a little but maybe some of it would be useful in the future...

I couldn't remember, so I did a quick search on Network Time Protocols and one reference noted "it is generally suitable for the synchronization of a corporate time server to an atomic clock on the global internet to within tens of milliseconds"

Now what we'd be interested in isn't synchronization with an internet server, but what typical syncronizations are like for low latency local networks.

I was thinking along the lines of two laptops synced over a 100mb crossover connection.

Then let them be linked over a wireless connection. Both record timestamps. One at start and one at finish. One can update the other, or you could have them both update a third that is setup on a local network connection to an access point. The third laptop doing nothing but recording and calculating the difference. Even if the syncronization is off between the two laptops...wouldn't they remain off by the same amount so that times would still be comparable?

The part that I'm not knowledgeable enough to work out is the part of going from the vehicle crossing the sensor to a sub-ms timestamp on the computer. (It would need to at least be consistent in latency.) And would have no idea about attempting to incorporate existing timing equipment...

especially that pretty display...

as for cost...low-end used laptops can be had for very cheap (especially after christmas on ebay)...and I would think something like this would make for a flexible and upgradeable system....but with everything involved, it'd easily break a $500 limit.

Hey, Mike, two words:

Quantum Entanglement

Sure, it's only been demonstrated in the laboratory so far, but it is absolutely instantaneous, with not even a speed-of-light lag. Even better, distance is no obstacle. The timing bus could be on Neptune and still see the spin state transition when a Miata breaks the light beam in Sanford. Better yet, the timing crew could be working from Hooters and just call in the times over a cell phone.

With a little DARPA funding, which I'm confident we can get, we might even make a little money on this. Could be patent rights in it, a Nobel for sure, and maybe even . . . dare I say it? . . . an appearance on Oprah. Think of it, Mike, national exposure for THSCC and it's farsighted prez.

I'll try to get prices for the liquid helium dewars and electromagnetic containment traps by the next meeting. By the way, can the alternator on the bus put out 50KV at a couple of hundred amps, or do I have to price a generator also?

Is there a prize for best first post ever?

Awesome!

_________________
Jim Pastorius
2008 Silverado VortecMax
1992 Camaro CMC#92
2002 BMW R1150R

2009 3rd Place CMC Mid-Atlantic Championship
2009 CMC Hyperfest Winner

Scary thing is, I actually understood like half of that. But could we put the bus on Uranus. Yeah, I had to go there.

_________________
www.ncvw.org
1993 VW Corrado VR6 - DSP15 (RIP)
1986 MB 190E 2.3L 16v - STX (SOLD)
1997 BMW M3 3.2L - BS for now

I might have an extra flux capacitor laying around.

_________________
SPIN or WIN!
there's no glory for going slow.

We girls have no interest in working from Hooters. So, when you set this thing up, you need to find some place else for the timing/scoring crew to work or you guys'll have to do all the work. -- Hey, maybe Hooters should be the place!

_________________
The person with too many names...
Mary E./ME/Emmie Fisher/Daniel/Daniel-Fisher
(& some others not suitable for posting!).
Help support our habits; BUY http://virginiabreeze.us !