Multi-Timer setup to capture performance data

General timing system discussions.
User avatar
pack529holycross
Master Pine Head
Master Pine Head
Posts: 555
Joined: Wed Dec 26, 2007 4:25 pm
Location: Dr. Phillips, Florida
Contact:

Multi-Timer setup to capture performance data

Post by pack529holycross »

I was thinking in my head uses for our former 3 lane timer hardware, and it occured to me that when watching races there are a few key points within the span of start - to - finish line that bear scrutiny when it comes to improving performance. Of course others may wish to add to this shortlist, but here is what I observed as "keypoints" in the race track:

1. 12-18" after starting pins ( showing strengths and weaknesses of certain cars when it comes to initial transfer of energy )

2. Middle of the transition ( showing relative quickness from starting pins to end of decline )

3. 4 - 8 ' from beginning of flats ( showing relative energy retained after transition )

4. 4 - 8' from finish line sensors ( showing relative energy retained after flat sections )


This brings me to ponder - if you added a phone jack splitter, would the remote switch on the starting gate trigger two timers? If so, an intermediate length timer could be set up at, say, 4 feet after curved section ends, to record times to that point. I believe this measurement would be helpful in gauging performance as you develop your cars. Also, it might be helpful to set up the timer at 32 feet on a 49 foot track to help develop a car that was strong at both distances, and not so severely disadvantaged on one length or the other because that length was ignored. We never know for sure what length of track will be used at an open or District event.

Any thoughts?
quadad
Master Pine Head
Master Pine Head
Posts: 494
Joined: Wed Jan 23, 2008 11:51 am
Location: SE, WI

Re: Multi-Timer setup to capture performance data

Post by quadad »

If I am getting your question right (about starting two timers), that's all up to the downstream wiring. Obviously if you have only one timer (as most designs do), you can only trigger that one and all timings will be the seconds elapsed from that event. I personally don't think this is all bad.

What I am considering as a future improvement to the test track I am currently building (that's scary :roll: ), is to be able to switch from timing three lanes in the standard manner, to four timings for one lane (whichever lane turns out to 'best' for car evaluation). In that manner I can either run a regular race or with just one timer get the kind of detailed information you are talking about here. I just would need to invest in the additional sensors.

For me swapping a connector to go from Race Mode to Single Lane Multi-Timing Mode would be easier and cheaper than a 2nd timing board.

My data will go into a PC anyway (from the MicroWizard K1 kit), and so I can do anything I want with the numbers.
quadad
Master Pine Head
Master Pine Head
Posts: 494
Joined: Wed Jan 23, 2008 11:51 am
Location: SE, WI

Re: Multi-Timer setup to capture performance data

Post by quadad »

Pack529holycross - not sure if you have done anything with this idea yet, but I am starting to plan to for my mid- to late-summer project. As posted in another forum, my latest thoughts are to switch from two different sensor sets and gather the following data for car performance measurement:

1) Pin drop to say ~1 foot down the track;
2) Pin drop to shortly before the transition;
3) Pin drop to ~ one foot after the transition;
4) (normal) Pin drop to finish line.

I see doing this with one timer, to save on cost and computer interface. Once you have these numbers in the computer, simple addition would allow for the interval times to be calculated.

I also am wary of getting surprised by our car times when we go from one track to another.
User avatar
SlartyBartFast
Master Pine Head
Master Pine Head
Posts: 272
Joined: Mon Feb 12, 2007 11:30 am
Location: Montreal, Quebec

Re: Multi-Timer setup to capture performance data

Post by SlartyBartFast »

This is exactly something I was pondering.

I was initially wondering if the timer at http://derbytimer.com/ would be able to be reprogrammed to log multiple inputs.

If each timing gate were a one-shot circuit, all gates on each lane could be wired to the same pin on the timer.

To take the whole thing a step further, I thought it would be cool to put two timing gates in close proximity at each point in order to be able to calculate car speed.

Unfortunately, my controller programming and electronics knowledge remains very limited at this stage.
quadad
Master Pine Head
Master Pine Head
Posts: 494
Joined: Wed Jan 23, 2008 11:51 am
Location: SE, WI

Re: Multi-Timer setup to capture performance data

Post by quadad »

SlartyBartFast wrote:If each timing gate were a one-shot circuit, all gates on each lane could be wired to the same pin on the timer. ...
Unfortunately, my controller programming and electronics knowledge remains very limited at this stage.
I think your idea is fine and anything is possible once you start playing with the programming. Even knowing how though, it takes some real time and means to program the PIC or other microcontroller to do a custom job. While I can do the work, I can't justify the time. It seems easier to me to just build a second set of IR LEDs and sensors and swap connectors when changing modes (and then doing the subtraction after the fact for time intervals).

Interestingly, the most expensive part of all of these kits is the display hardware (if you want that). I choose to go with a PC-interface based kit (cheap K3) from Microwizard:
http://www.microwizard.com/k1cheap.html

Price might be comparable - I didn't check in detail, but their support is fantastic. I was surprised that the derbytimer had most everything on Backorder, with this being the off-season (as far as I can tell). You might want to contact someone there though, because the smallest outfit might be the one most willing to do a custom programming job for you - just a thought. I'm sure others would be interested if this was available.
User avatar
gpraceman
Site Admin
Site Admin
Posts: 4926
Joined: Fri Jun 20, 2003 12:46 am
Location: Highlands Ranch, CO
Contact:

Re: Multi-Timer setup to capture performance data

Post by gpraceman »

quadad wrote:Interestingly, the most expensive part of all of these kits is the display hardware (if you want that). I choose to go with a PC-interface based kit (cheap K3) from Microwizard:
http://www.microwizard.com/k1cheap.html
Looks like you mean the $60 K1 "cheap kit".
quadad wrote:I was surprised that the derbytimer had most everything on Backorder, with this being the off-season (as far as I can tell). You might want to contact someone there though, because the smallest outfit might be the one most willing to do a custom programming job for you - just a thought. I'm sure others would be interested if this was available.
He's had everything listed as on backorder during the entire 2007-2008 season. Bert Drake also has not been responding to emails.
Randy Lisano
Romans 5:8

Awana Grand Prix and Pinewood Derby racing - Where a child, an adult and a small block of wood combine for a lot of fun and memories.
quadad
Master Pine Head
Master Pine Head
Posts: 494
Joined: Wed Jan 23, 2008 11:51 am
Location: SE, WI

Re: Multi-Timer setup to capture performance data

Post by quadad »

gpraceman wrote:Looks like you mean the $60 K1 "cheap kit".
Yep, your right. Thanks Randy.

I wanted to add one other thing from my point of view. If you are going to do this 'hard core' interval timing, it really only makes sense to me to do it on one given lane of any track and to concentrate on one car at a time. That way there are no "lane-to-lane" differences in the data for different cars. Its just one more reason why I favor the 'swap the connector' approach. I can understand others having different interests though.
User avatar
SlartyBartFast
Master Pine Head
Master Pine Head
Posts: 272
Joined: Mon Feb 12, 2007 11:30 am
Location: Montreal, Quebec

Re: Multi-Timer setup to capture performance data

Post by SlartyBartFast »

gpraceman wrote:He's had everything listed as on backorder during the entire 2007-2008 season. Bert Drake also has not been responding to emails.
Yeah, the forums seem dead too. I tried to join the Yahoo group, will see if I get a response from the moderator (I'm assuming that's Bert too).

But, it's quite simple to put the schematics for the derbytimer into EagleCad and then send off the file to have a single board produced. The source code for the timer is also on the website.

I purchased a DIY AVR programmer kit with the intention of porting the derby timer to AVR from PIC and attempting to implement a few of my crazy ideas.

Some not so crazy improvements were the addition of sensor sensitivity adjustment and lane status indicators.

damned annoying thing is I had done most of the circuit drafting when my portable hard drive failed on me. Now it's a paperweight that buzzes and clicks if you plug it in. :shake:

I had the intention of copying the NASCAR style position tree. A vertical stack of 7-segment displays that would display the finish order from top to bottom.

Just seems to me that 3, 4, or 5" LED displays are more impressive than a projected image.

My HUGE concern with the idea of timing performance is the issue of clock speed and signal timing. How long to make the one-shot of each sensor, whether to use two sensor inputs per interval or combining the two on one controller input (one for interval time, second for speed). Makes for lots of signals as 6 cars go through 12 sensors and you want to accurately time all of them.

Then what distances can signals travel? I2C to RS232 chips can be used to boost signal strength, but with what effect on timing?

I had thought of perhaps designing a tiny timer that would time one lane but be able to communicate using serial and be able to be cascaded. Or a single interval timer with finish and speed sensors that could be cascaded for multiple intervals.

I really need an electronics expert to guide me. Too many ideas, too many concerns, too little knowledge...
quadad
Master Pine Head
Master Pine Head
Posts: 494
Joined: Wed Jan 23, 2008 11:51 am
Location: SE, WI

Re: Multi-Timer setup to capture performance data

Post by quadad »

SlartyBart,

Are you talking about a 6-lane track with multiple (how many ?) sensors for each lane ? Do you want to race simultaneous with getting "interval timing" from these ? Please clarify what you all want to accomplish. Note that the position tree stuff is all a separate issue from the timing.

In general, there is no rocket science to determining the acceptability of your chosen timing approach if you have decent data sheets. Are you trying for beyond-the-norm levels of timing resolution or accuracy ? Do you believe that today's existing timers are inaccurate or ? - just wondering.

Sending digital signals 10's of feet is no bid deal and doesn't require 'boosting'. RS-232 flavored transmission is OK, but not what I would call 'industrial strength' if you are worried about running this next to say an arc welder.

I am not a PIC expert, but would think that sufficient inputs and outputs are part of the issue with the lower end devices. Again, this goes back to my first question above.

If you aren't in a world of hurry, feel free to PM me with more details of what you are considering electronics wise. A block diagram is a good place to start. I may be able to help you, but not too quickly.
User avatar
SlartyBartFast
Master Pine Head
Master Pine Head
Posts: 272
Joined: Mon Feb 12, 2007 11:30 am
Location: Montreal, Quebec

Re: Multi-Timer setup to capture performance data

Post by SlartyBartFast »

quadad wrote:I may be able to help you, but not too quickly.
Well, I'm obviously not on the forum for speed. :P

My original idea was to take the timer from DIYTimer and reprogram it. Current timers are plenty acurate, just not as much data as I imagine we could get. :)

Racing would be simultaneous on all six lanes. There would be multiple intervals (or timing gates) on the track.

For each interval, each lane would have two detectors. The first detector would be used for interval time, the second for calculating speed.

I'm figuring the setup could easily be modular. A processor module to which timing gate modules would be wired. Either the timer is configured to indicate the number of intervals or, the final timing gate connects to a different set of inputs so the timer knows which is the final overall time.

I had the whole thing block diagrammed and even had schematics an the beginnings of boards on EagleCAD. Then my portable drive died. :(

But, I'm stabbing in the dark really. It's been a long time since my university electronics courses. But, like programming it seems you can "borrow" lots of predesigned blocks.

A copy of many of the fine features in Paul Taylors setup would be a great design goal .http://derbytalk.com/viewtopic.php?p=36635#36635[/url]
knotthed
Master Pine Head
Master Pine Head
Posts: 169
Joined: Tue Dec 13, 2011 9:01 pm
Location: Northeast, Illinois

Re: Multi-Timer setup to capture performance data

Post by knotthed »

Did or does anyone have a "Multi-Timer setup to capture performance data"?

If you do, do you feel that it helped you test and tune cars?

Thinking of doing the same thing, but would love to hear some lessons learned first.
User avatar
psycaz
Master Pine Head
Master Pine Head
Posts: 667
Joined: Wed Apr 13, 2005 9:21 am
Location: Somewhere, US

Re: Multi-Timer setup to capture performance data

Post by psycaz »

I had a muti-point timing system on when I was using a wooden track for testing/tuning.

I thought it helped me tremendously.

It made it simple to tell if it bumped the track, triggered the start gate differently, those types of issues in particular.

Without it, I just would have thought it was my tuning sometimes. As it was, I could see the differences to the interval (bottom of the ramp) and see the time differences.

Did also help with figuring out how the car wanted to stage as I could watch the times to that first interval and see if what I was doing was making a difference.
knotthed
Master Pine Head
Master Pine Head
Posts: 169
Joined: Tue Dec 13, 2011 9:01 pm
Location: Northeast, Illinois

Re: Multi-Timer setup to capture performance data

Post by knotthed »

Can you elaborate more on your particular setup?

How many timing points and locations?
User avatar
psycaz
Master Pine Head
Master Pine Head
Posts: 667
Joined: Wed Apr 13, 2005 9:21 am
Location: Somewhere, US

Re: Multi-Timer setup to capture performance data

Post by psycaz »

I have a 32' wooden track. Two lanes, one of which I mainly used for tuning. 3 timing points as that is what came with my DT1000? timer :). First at the bottom of the ramp, second at the 24' mark, and the last at the finish line.

I used the last two to get what I could for how much the cat was slowing. The first to check gate to that point.

Track has a manual release. That first timer quickly showed how much I was influencing the times just with the way I was releasing the gate. so I got better there real fast. As I said, next was if I was staging incorrectly. I could tweak the time several thousandths easily without ever touching anything on the car and not thinking I had changed the stage.
knotthed
Master Pine Head
Master Pine Head
Posts: 169
Joined: Tue Dec 13, 2011 9:01 pm
Location: Northeast, Illinois

Re: Multi-Timer setup to capture performance data

Post by knotthed »

Do you feel that 3 points is enough or If you could have had more check points, where would you have put them?

I am thinking one closer to the gate, say within the first 2 feet of the pin.

before the curve, after the curve, then the flat followed by final.

Maybe 5?

I guess the real question is.......is 5 better than 3? Would it help to have the couple extra timing points?
Post Reply