entropy
Moderator
Posts: 8671
|
posted August 18, 2004 05:19 AM
electronics expert needed: 5v digital to analog
OK, here is a chance for the electronics experts to shine!!!
I have just installed the most-excellent (and popularly priced) Innovate LM-1 A/F data logger on my drag bike. WORKS GREAT!
On the aux logging channels, it is only able to take a 0 - 5v input signal and with software you calibrate that range to what you are monitoring. Like throttle position, 0% to 100% throttle opening.
I wanna monitor r wheel speed, but wheel speed sensors put out a digital 5v signal so I need a popularly priced analog to digital counter which outputs a 0 - 5v signal which is linear w/the pulses/sec coming from the wheel speed sensor.
If you have one or can make one of these you could prolly sell it to Innovate for reselling with the LM-1, because they currently don't offer wheel speed monitoring.
Sparky????  I know you are out there with a solution!
____________
This moderator uses moderation in moderation
  
|
frEEk
Administrator
ummm... yeah
Posts: 9660
|
posted August 18, 2004 08:30 AM
check Radio Shack (do u have those in the US?). they almsot certainly have that. it's a rather common thing to do after all. u basically jsut need the right size capacitor and resistor if i remember my four years of wasted education correctly.
|
entropy
Moderator
Posts: 8671
|
posted August 18, 2004 09:18 AM
thx FrEEk,
"if i remember my four years of wasted education correctly." bahahahahaha!
my education is also wasted (as far as dis-remembering things); prolly because I was wasted. Radio shack is a good tip!
____________
This moderator uses moderation in moderation
|
ridgeracer
Pro
Posts: 1309
|
posted August 18, 2004 02:59 PM
 Edited By: ridgeracer on 18 Aug 2004 16:17
Doesn't Innovative make an 'RPM cable' for the LM-1? The ZX-12 already generates a digital rear wheel speed signal that could be connected and logged the same as an RPM signal.
|
slug
Pro
Out in search of my mind...
Posts: 1433
|
posted August 18, 2004 06:50 PM
Edited By: slug on 18 Aug 2004 19:52
you'll need to determine the pulserate, to get the actual conversion
here is general selection table for analog devices:
http://www.analog.com/Analog_Root/sitePage/mainSectionContent/0,2132,level4%253D%25252D1%2526ContentID%253D18132%2526Language%253DEnglish%2526level1%253D117%2526level2%253D128%2526level3%253D%25252D1,00.html
try also onsemi, i have had VERY good dealings with their customer support.
be armed with the pulse rate of the 5V pulse train, and the output voltage range that you wish to see (0-5 V 0=0mph, 5v = 190mph, or whatever.)
they have devices that will convert just about anything.
OnSemi's datasheets are VERY complete as far as describing the circuit usage, so you can set it up properly in the circuit.
http://www.onsemi.com/site/products/summary/0,4450,NE5008,00.html
but with the correct information upfront, it shouldn;t be too hard to make a converter box to allow that connection. even throw in some opto-isolators to protect the expensive bits from damage if the converter box should fail, get struck by lightning, crash on the track, etc
  
|
fish_antlers
Administrator
The Truth is Out There
Posts: 21894
|
posted August 18, 2004 09:43 PM
you need to ping ridgeracer.... he'll be able to help ya
____________
What business is it of yours where I'm from, Friendo?
|
entropy
Moderator
Posts: 8671
|
posted August 19, 2004 03:02 AM
quote:
Doesn't Innovative make an 'RPM cable' for the LM-1? The ZX-12 already generates a digital rear wheel speed signal that could be connected and logged the same as an RPM signal.
Ridgeracer,
They do make an rpm cable (i've got one), it has rpm input (digital), and 4 analog input terminals. The LM-1 box only takes one rpm cable.
If i wanted to sacrifice rpm in favor for wheel speed, i could hook it up to the 12's system, but since the 12's wheel speed is calculated from rpm and gearing, i wouldn't know if the wheel was slipping or the clutch was slipping.
I THINK i really need to put a sensor on the r wheel.
1. the wheel sensor would put out about 45 pulses/sec at 200mph
2. the LM-1 can take a max of 5v (analog)input;
3. bike has 12v avail to power the DtoA box and the wheel sensor;
I guess those are my "functional specs"
____________
This moderator uses moderation in moderation
|
entropy
Moderator
Posts: 8671
|
posted August 19, 2004 03:10 AM
Slug;
thx for the links, and they sure have a lot of stuff. I bet you even know how to read those data sheets, eh?
How about I contract this job out to YOU!
____________
This moderator uses moderation in moderation
|
ridgeracer
Pro
Posts: 1309
|
posted August 19, 2004 06:36 AM
Edited By: ridgeracer on 19 Aug 2004 08:52
The 12's speed pulse is triggered off the teeth of gear on the output shaft which is on the wheel side of the clutch, not the crank side. It does give you an accurate rear wheel spin, untill you snap your drive chain anyway  This is the same signal that drives the speedo. If you pull in the clutch and rev the engine your speedo doesn't move does it? Besides I don't think there is any way to measure rear wheel slip which would be an increase in rear wheel rotation with out an increase in ground speed. What are you going to compare your wheel spin to? How are you going to measure your true instantaneous ground speed? All a wheel spin input will tell you is how well your clutch is working. Another wheel spin sensor isn't going to tell you anything you can't already learn from the existing speedo sensor.
If I remember right I think the 12 generates 38 pulses per wheel rotation or 1083 Hz (pulses per second) at 60 mph.
BTW Slug...how do you input a 5V pulse train into an 8-bit parallel load DAC?
|
entropy
Moderator
Posts: 8671
|
posted August 19, 2004 07:48 AM
quote:
The 12's speed pulse is triggered off the teeth of gear on the output shaft which is on the wheel side of the clutch, not the crank side. It does give you an accurate rear wheel spin, untill you snap your drive chain anyway
ridgeracer,
good clarification, I didn't know that! So I DON'T need a wheel sensor, just the DtoA box.
Do you know whether the signal which drives the speedo is digital or analog? I'm guessing digital, but wouldn't it be convenient if it were analog!
Thanks for teaching an old dog a new trick
____________
This moderator uses moderation in moderation
|
slug
Pro
Out in search of my mind...
Posts: 1433
|
posted August 19, 2004 04:20 PM
ridgeracer: what i'd do is call onsemi and ask them,with up to whatever frquency (5 kHz prolly be enough 300 MPH?) digital pulse train the best DAC to use to get a 0-5V analog output, with a 12V supply voltage
as for the other, again i'd have to go to the datasheets to be sure onthe actual connections.
i kinda just posted a few links to a few sites i have used at my factory job to find parts. but most cases a call to customer support was used to get the actual details.
if i get home early enough tomorow i'll call onsemi and ask them
just let me know what the top speed you want indicated is so i can get the right spread (or a variable spread if possible)
|
entropy
Moderator
Posts: 8671
|
posted August 19, 2004 04:55 PM
Edited By: entropy on 20 Aug 2004 01:55
slug,
0 -220 mph oughta cover all situations.
I REALLY appreciate any help you can drum up!!!!
:thumbsup:
PS: i'm going out for some authentic Chinese food, and will be in spotty e-connection for a week, but I am VERY interested in getting a solution here!!!!
____________
This moderator uses moderation in moderation
|
ridgeracer
Pro
Posts: 1309
|
posted August 20, 2004 05:37 AM
Edited By: ridgeracer on 20 Aug 2004 06:45
Entropy,
Unfortunatley it is a 5V digital pulse. Also you are going to have a hard time finding an off the shelf 'DtoA' box.
While there are thousands of devices that will convert voltages to numbers (ADC) and numbers to voltages (DAC) it is not common to need to convert pulses to voltages. Actually with a pulse what you are measuring is the time between pulses. So you really want to convert time to voltage.
Your going to need a two stage device consisting of some kind of small single chip micro to measure the pulse and convert it to a number which it passes to a DAC which will convert the number to a voltage.
Freek's idea of using a cap and resistor would be cheap and technically work but it would be very non-linear. A cap that took say 5 seconds to fully charge would go from 0% to 66% in the first second while the charge would only change 3% during the last second. Also the cap is actually responding more to pulse width than time between pulses.
Now thinking outside the box for a moment, there may be a software solution to your problem. Your data logger just generates a bunch of numbers that I assume you use some windows software to download and view in graphical form. We could log the data in a non standard way and then massage the data into something the windows software, and you, could understand. Let me give you some examples.
First just feed the speedo input directly into a spare analog channel. The logger would record just a bunch of alternating very high and very low voltages. On the graph it would just look like a square wave. But you would notice that as the bike went faster the high parts and low parts get smaller and move closer together. You would actually have all the data needed to derive wheel speed. It just wouldn't be in a user friendly format. But you could, well maybe you couldn't, but someone could then right a simple program to look at your wheel channel data, measure the time between pulses, convert it to a number and rewrite it to the data file as numbers that when viewed look like an analog value. So you would log it during a run, download it to a pc, convert it with your new program, then view it. The only problem here might be sample rate of your logger, how many times it records a data set per second.
Another option would be to multiplex your RPM channel. Since the RPM channel is designed to measure pulses, feed both your RPM and wheel speed to the same channel but alternate between them many times a second. You would put a simple switching device that would show RPM one sample and wheel the next sample. As before the graph would be hard to read but all the info would be there. You would run it through a converter program that would take your odd samples and display them as rpm and your even samples as wheel speed. The missing samples would be averaged in for each channel. So if you had 5036 rpm in sample 1 and 5038 rpm for sample 3 the program would just assume 5037 rpm for missing sample 2.
I'll think about it some more over the weekend. And check out the specs of the LM-1. I like the software idea because it requires little or no extra hardware and you could easily email the program to anybody who wanted to use it.
|
ridgeracer
Pro
Posts: 1309
|
posted August 20, 2004 10:07 AM
Edited By: ridgeracer on 20 Aug 2004 11:22
Ok, Contrary to what I said above I actually found a part that may do what you need.
http://cache.national.com/ds/LM/LM2907.pdf
It is a Frequency-To-Voltage converter that puts out a variable voltage depending on input frequency. You can also use it as a frequency switch where it turns on above a set frequency (Think shift light)
Of course it will set you back $2.05. from http://www.digikey.com The actual part you want is LM2917-ND. This is a larger 8 pin DIP part that will be easier to work with. If you sneeze you'll blow the surface mount parts across the room. Not to mention trying to solder little pins that are .050 inches apart
Actually if you go to National Semiconductors site they would probably 'sample' you one or two for free. But it may take weeks to get
I just might be tempted to build up one to see how well it works, not just for your project but for a shift light and/or rpm limiter. Would you be able to build one from a schematic and parts list?
|
spencercyclecom
Zone Head
Posts: 535
|
posted August 20, 2004 08:13 PM
checxk this out...Carl.
I just got a email form them alerting me. How about 6 MORE items to datalog!???
Just went to production
The LMA-3, also known as the AuxBox, is released from engineering. This device attaches to the auxiliary input of the LM-1 and allows a user to log and analyze 5 critical engine metrics in addition to the air/fuel ratio native on the LM-1. The AuxBox makes the LM-1 into a complete tuning kit, allowing the user to capture, view, and analyze data on all 6 channels that the LM-1 can log internally. Correlating AFR, RPM, MAP, and CHT, for example, is essential when tuning piggy-back fuel controllers, ECU programmers, aftermarket ECUs, or racing carburetors.
The 5 internal sensors are intended for capturing Manifold Air Pressure (MAP or Boost), Cylinder Head Temperature (CHT) or Exhaust Gas Temperature (EGT), RPM conversion (from a tach signal or inductive clamp), Acceleration, and Injector Duty Cycle (or Dwell). The internal MAP sensor is a 3-bar absolute pressure sensor, providing accurate data up to 44.1 PSIA. The temperature measurement circuit features a K-type thermocouple. The internal accelerometer features 2 axis measurements for a variety of mounting options.
Another key application of the AuxBox is road testing. Even a dyno doesn't provide as much real data as road testing. Factors like air flow, side force, and load vary significantly on a real road. With an AuxBox, a user can log 6 channels, in multiple sessions, for up to 44 minutes. Then, back at the garage, simply connect the LM-1 to you PC and view all of you log data. Use LogWorks to view fuel mapping, perform statistical analysis, overlay graphs, and playback a log on the virtual gauges. You can also view realtime data on LogWorks' gauges if you want to take your laptop on the road.
The AuxBox also includes 5 external inputs for connecting external sensors in lieu of the internal sensors. This allows users to log data from external sensors like TPS or suspension data. Each input can also be configured with additional thermocouples, allowing multiple EGT channels.
The AuxBox ships with mounting hardware, a 6' thermocouple wire, the LM-1 connecting cable, and the user manual. Retail. Price is $249.
____________
Thanks
Doug Ray
SpencerCycle.com
SpencerCycle Center Inc.
Spencer Performance Group Inc.
Toll Free 1-800-510-1945 ext #10
International 1-800-510-1945 ext #11
Local/Tech 1-704-642-0987 ext #12
Fax 1-704-642-0987
|
entropy
Moderator
Posts: 8671
|
posted August 20, 2004 09:18 PM
ridgeracer,
i could prolly work with a schematic & parts list, but i'd be happy to contract YOU, an expert, to put it together
I'm traveling until next Thurs, but will try to keep in e-connection; it'll depend on my Shekou & Beijing hotel connections.
Let's build this thing!!! You make a few bux from me and you keep the intellectual rights to the product. Yep, sell it to innovate!
____________
This moderator uses moderation in moderation
|
entropy
Moderator
Posts: 8671
|
posted August 20, 2004 09:19 PM
Doug,
that is a cool box! I'm gonna get my base unit running right, then expand.
Thx for the update,
Karl
____________
This moderator uses moderation in moderation
|
ridgeracer
Pro
Posts: 1309
|
posted August 21, 2004 05:52 AM
It sounds like the LM-3 still only has one digital (RPM) input.
Innovative would never pay me for the design because they could solve this problem with a very simple design change on their part. Of course there are probably a few, like you, who already have an LM-1 who would like a wheelspin / speedo option.
I will go ahead and design and build a few prototypes. Entropy can be my beta tester. Then I will post all the details on the web somewhere like I did with the 12r speed derestrictor and gear indicator. I've always been more of a Linux open source type when it comes to intellectual property.
Its all about making the world a better place and if I can help at least one person somewhere in this big world of ours go faster on a ZX-12....well thats all the pay I need. I'll be able to look back in my old age and say..." I made a difference..."
Besides you couldn't afford to 'contract' me to do it for time and materials.
|
slug
Pro
Out in search of my mind...
Posts: 1433
|
posted August 21, 2004 06:59 AM
wow, and i am such a slug for not getting to call onsemi yesterday..
ridge: glad you found a high-dollar part to solve the question
heheh
i knew there were parts out there to do that,just a matter of finding them. (several pieces of equipment in factory did exactly that, converted pulse train to an analog signal for the processor)
anyway glad we have apart to start with. looks like it even has some sample circuits that could bemassaged to work correctly.
|
entropy
Moderator
Posts: 8671
|
posted August 21, 2004 04:57 PM
quote:
...
I will go ahead and design and build a few prototypes. Entropy can be my beta tester.
... I'll be able to look back in my old age and say..." I made a difference..."
woohoo!!!
a cool lil development project!
____________
This moderator uses moderation in moderation
|
NINJA12
Needs a job
Posts: 3310
|
posted August 25, 2004 01:01 PM
keep this close to the top!
|
entropy
Moderator
Posts: 8671
|
posted August 28, 2004 12:16 AM
now i have the tps, rpm and AF all recording perfectly.
rear wheel mph will be the icing on the cake!
ridgeracer, you DA MAN!!!
____________
This moderator uses moderation in moderation
|
ninja12
Needs a job
Posts: 3310
|
posted August 28, 2004 03:04 PM
Karl, what did you have to do to make it all work?
That's my main objective, to build my own maps.
I don't want any suprises AFTER i get this thing.
Ridgeracer if you get this to work would you build me one also?
I'd be happy to pay.
|
entropy
Moderator
Posts: 8671
|
posted August 29, 2004 06:47 AM
Edited By: entropy on 30 Aug 2004 05:05
Ninja12
to make it work was really a piece of cake.
to get AF
1. weld the bung on yr mid pipe,
2. mount the LM-1 box somewhere you can get at it; mine is on the lh ram air cover
3. connect the AF sensor to the bung, connect cable to LM-1 box
4. get power & ground for the LM-1 box
5. 2 min calibration of the AF,
6. done - you are measuring and recording AF
to add rpm
1. plug the rpm module into the LM-1 box, i put the rpm module in the trunk
2. take rpm signal from a wire going into the ECU, connect to rpm module
3. take gnd from the br/bk wire going into the ECU, connect to rpm module
4. fire up and calibrate the rpm and scale
5. done - you've got rpm (and AF)
to add TPS
1. take tps signal from a wire going into the ECU, connect to rpm module
2. set up tps parameters in the sw
3. done - you've got tps (and AF and rpm).
I made 6 passes last night at the Texas shootout (woo-hoo, 4 rounds), it recorded perfectly
To make map changes?
Johnny Cheese was modifying my PC3 settings by looking at the AF graph after a pass and mph off the ticket; AF changes about .2 per bar; I gained 1.5mph in the 1/8 with the changes. i was racing SET so 1/4 mph wasn't used (backing off).
It would be even better to actually go into the map and make changes, but we were really just experimenting.
____________
This moderator uses moderation in moderation
|
Ra12r
Zone Head
Posts: 919
|
posted August 30, 2004 05:14 AM
Hmmmm, would you possibly be getting "Optimal Performance" tuning without a 100% stress test of your bike ???
|
|
|
HOME
NEW TOPIC