VTTOM, sent you an email with info as requested.
VTTOM, I have sent you an email. Please respond and I can tell you what I have done so far.
I think I've found the perfect microcontroller for this project. The PICAXE-08. It's got 4 general-pupose inputs and serial in/out. It doesn't even require an external crystal. I should be able to power it using a 5v regulator off the S149's printer batteries (I plan to embed it inside the body of the timer and put my own 3-pin serial interface on it).
Reading the 4-bit bus should be pretty easy. The encoding they've chosen means that no 2 sequential characters can be the same. So, I simply sample the data and compare it against the previous sample. If it's the same, do nothing. If it's different, then I know I've got the next nibble. From there, it's a simple matter of translating the proprietary character code to ASCII and then sending it out the serial port.
What do you think?
Reading the 4-bit bus should be pretty easy. The encoding they've chosen means that no 2 sequential characters can be the same. So, I simply sample the data and compare it against the previous sample. If it's the same, do nothing. If it's different, then I know I've got the next nibble. From there, it's a simple matter of translating the proprietary character code to ASCII and then sending it out the serial port.
What do you think?
Wow! This is a very nice chip. Your idea of putting the chip into the stopwatch is a very good one. I am not sure that I am prepared to do that just yet. I like the idea of keeping the stopwatch in it's original state. I also want to incorporate multiple channels into my design so I think I will try the 18F2420 for now. Although it would be easy to chain a couple of these for more ports....
I do have to buy some of these though looks very easy to program and use.
Sounds like you are going to try the same logic I used with the Parallax chip. It should work just fine.
Keep me posted.
Luc..
I do have to buy some of these though looks very easy to program and use.
Sounds like you are going to try the same logic I used with the Parallax chip. It should work just fine.
Keep me posted.
Luc..
I'm beginning to think that perhaps I should make this work as an external dongle instead of putting it inside. That way, I'd have something I could share with other groups who might want to do the same thing. Problem is, that connector looks non-standard. Anyone know of a source for the 8-pin data connector?
Time to scoure the Internet for a source.
Time to scoure the Internet for a source.
connector is non-standard & I looked on the net but had no joy.
If you look an page one I think Luc mentions he turned some up on a lathe.
G
If you look an page one I think Luc mentions he turned some up on a lathe.
G
Yes, I have made one already. It's not to difficult. The biggest problem is getting a female pin small enough for the Seiko. I am just using some wire right now (hard to explain), but it works. I have some ideas on how to make them in bulk but won't start until I can get the interface to work.
Luc..
Luc..
OK. I ordered a 1/2 dozen PICAXE-08's. I'll let you know how it goes after they arrive and I start programming with them.
I recevied a PICAXE microcontroller last week, and I've been able to hook it up to my S149 and started to play around with observing the data on he 4-bit bus. However, 2 things don't quite match my expectations:
1) I think I'm seeing at least one of the bits on the data bus (DB3, the MSB) wiggle even when there is no data to be sent. Do I have a flaky connection or is this expected behavior?
2) My first impression was that BUSY is an output from the timer. Now I'm beginning to think it's an input TO the timer, to be used for handshaking between the timer and the out-board serializer I'm building.
Can anyone offer some clarification here?
1) I think I'm seeing at least one of the bits on the data bus (DB3, the MSB) wiggle even when there is no data to be sent. Do I have a flaky connection or is this expected behavior?
2) My first impression was that BUSY is an output from the timer. Now I'm beginning to think it's an input TO the timer, to be used for handshaking between the timer and the out-board serializer I'm building.
Can anyone offer some clarification here?
Nope, not a flaky connection. I was getting the same thing with the Parallax and PICAXE as well. I changed my logic to only start reading on a value higher than 9 (looking for an A,B,C,D) before I read the other bits.
I suspect that there is other data on the bus most of the time and the printer/port data is only relevant when the busy signal is active. Initialy I thought that I could write some logic to ignore the busy signal but maybe not. So yes I believe it is to do with the handshaking for the output.
The funny part is that the busy signal sits at 1.5 volts most of the time and then goes high and then low, or low then high,can't really remember right now. But I think understanding and reading the busy signal is the clue. I haven't tried it recently but will try it again in the next few days.
Luc..
I suspect that there is other data on the bus most of the time and the printer/port data is only relevant when the busy signal is active. Initialy I thought that I could write some logic to ignore the busy signal but maybe not. So yes I believe it is to do with the handshaking for the output.
The funny part is that the busy signal sits at 1.5 volts most of the time and then goes high and then low, or low then high,can't really remember right now. But I think understanding and reading the busy signal is the clue. I haven't tried it recently but will try it again in the next few days.
Luc..
I have discovered by playing around with a jumper wire and a 10k pull-up resistor connected to the BUSY signal, that if I pull it high, and then hit the split/lap button a bunch of times, it keeps the printer from printing until I yank the resistor and let BUSY float, at which point the printer dumps the accrued lines of data.
So, it looks to me like the BUSY signal is a "wired OR", providing a means for telling the data interface to buffer the data. So, I'm thinking that so long as your microcontroller is fast enough, it shouldn't have to observe or control BUSY.
So, it looks to me like the BUSY signal is a "wired OR", providing a means for telling the data interface to buffer the data. So, I'm thinking that so long as your microcontroller is fast enough, it shouldn't have to observe or control BUSY.
Nice work. I spent a couple of hours last night working through some logic with the PICAXE. Looks like this chip is not fast enough. I am able to read the correct values (a very good thing) if I do a continuous read and write to a varable then display the variables. It only reads each change once or twice and if I add some logic to compare the previous value I miss alot of the data. So the logic to look for an "A" as the start of the data stream seems to work well. I just need a faster chip now.
I don't have a chip burner anymore I had to return the one I had. I have purchased a new one and waiting for delivery. I will try to use the Microchip 18F2520 when I get the burner.
Luc..
I don't have a chip burner anymore I had to return the one I had. I have purchased a new one and waiting for delivery. I will try to use the Microchip 18F2520 when I get the burner.
Luc..
> Looks like this chip is not fast enough.
Unfortunately, I more-or-less came to that same conclusion myself last night. However, I'm going to continue to play with it a little, to see if I can't find something that works. For instance, instead of sampling the data inside a loop, I'm hoping that maybe I can use one of the higher-level bit-blasting BASIC commands.
I also could go back to a discrete implementation that I'd sketched out earlier.
Also, someone suggested I try using custom software on a PC to sample the data using the parallel port. I know there is a bunch of freeware that lets you use your parallel port as a poor man's logic analyze. I could try going that route as well.
Unfortunately, I more-or-less came to that same conclusion myself last night. However, I'm going to continue to play with it a little, to see if I can't find something that works. For instance, instead of sampling the data inside a loop, I'm hoping that maybe I can use one of the higher-level bit-blasting BASIC commands.
I also could go back to a discrete implementation that I'd sketched out earlier.
Also, someone suggested I try using custom software on a PC to sample the data using the parallel port. I know there is a bunch of freeware that lets you use your parallel port as a poor man's logic analyze. I could try going that route as well.
BTW - Googling for "parallel port logic analyzer" turned up the following, which looks really usefull. I'm going to give this a try to better understand just exactly what this little 4-bit (+ BUSY) bus is doing, and how fast.
http://www.xs4all.nl/~jwasys/old/diy2.html
http://www.xs4all.nl/~jwasys/old/diy2.html
So I think I managed to get the PICAXE-08M working after all. I am getting spurious bad data, so I'll have to look into that. I also haven't tried a full recall, either, so I might need to make some alterations there, as well.
In principal, what it does is monitor the bus for a starting value of 0xA (0xF doesn't work for some reason) and then captures 26 nibbles (1 nibble = 1/2 byte = 4 bits), stuffing them into a buffer. After that, it then goes back and does the more time-intensive task of pairing up the nibbles into characters, doing the translation to ASCII, and then transmitting the data over a serial port.
In principal, what it does is monitor the bus for a starting value of 0xA (0xF doesn't work for some reason) and then captures 26 nibbles (1 nibble = 1/2 byte = 4 bits), stuffing them into a buffer. After that, it then goes back and does the more time-intensive task of pairing up the nibbles into characters, doing the translation to ASCII, and then transmitting the data over a serial port.
Hi all,
not sure about all the technical stuff your talking about but i have found the following using the genuine unit-
With CPS plus you can filter unwanted bits of info, you can substitute bits for different bits(ie a space for a .) good for text to column in excel also remove the last 15 bits(the split which appears even with button off). it also needs a baude rate of 1200, no parity, 8 bits and xonxoff.
Its worth getting it to work as we are getting great results.
Rod
not sure about all the technical stuff your talking about but i have found the following using the genuine unit-
With CPS plus you can filter unwanted bits of info, you can substitute bits for different bits(ie a space for a .) good for text to column in excel also remove the last 15 bits(the split which appears even with button off). it also needs a baude rate of 1200, no parity, 8 bits and xonxoff.
Its worth getting it to work as we are getting great results.
Rod
I am now working on a prototype now using a PIC18F2520 chip. This chip is very fast and I am able to read all the data without a problem. I just need to clean up and comment my code and I will be able to publish some of the detail. I also hope to get the bootloader function working so the chip can be reprogrammed without a burner. Got my fingers crossed.
I have also purchased the required hardware for one unit. I will use this as the prototype for now and will order more after I have some others test the interface. I am going to code the output to the same as the Seiko printout for now but I can change it to almost anything. My goal is to have the same output as a Chronomix Timing unit so we can use our software to record the time.
The connector is probably the biggest problem right now. I have made one for testing but I would like to buid a cleaner unit for production. I am having problems locating the correct size pin.
Luc..
I have also purchased the required hardware for one unit. I will use this as the prototype for now and will order more after I have some others test the interface. I am going to code the output to the same as the Seiko printout for now but I can change it to almost anything. My goal is to have the same output as a Chronomix Timing unit so we can use our software to record the time.
The connector is probably the biggest problem right now. I have made one for testing but I would like to buid a cleaner unit for production. I am having problems locating the correct size pin.
Luc..
Let me know what you come up with in terms of a connector. I've had zero luck finding one from a parts supplier like Mouser or Digi-Key. I'm afraid even if I found one, they would be so expensive I wouldn't want to buy them. Unless things change, I'm pretty much resolved to opening the timer, installing my microcontroller inside, and then adding my own, non-proprietary connector for serial communications.
Hi! I have been trying to get info on this for the last few years. I help out with the high school cross country races, and have written a race management program in Excel/VBasic. It works really well at entering finishers in the chute, but the times have to be entered later manually from the S129 printout. It would be SO nice if the machine could "dump" the race times to my computer, and I could paste them into the results file. I am not that "techno" on the hardware, though - does anyone have anything available that works that I could just plug in and download the times?
Hi Cathy, yup its possible.
I have my setup so the timer dumps to a text file then Access using vba imports from the text file.
the tricky bit is buying the interface (as per the discussion thus far) I figure your best bet is to try and contact Luc (or possibly rod to see if Harold wood has any left but £££).
I have my setup so the timer dumps to a text file then Access using vba imports from the text file.
the tricky bit is buying the interface (as per the discussion thus far) I figure your best bet is to try and contact Luc (or possibly rod to see if Harold wood has any left but £££).