View Full Version : Inertia Dynamometer
08-30-2008, 05:54 AM
··I am trying to incorporate·the Basic Stamp into a motorcycle inertia dynamometer. The system will need to measure drum revolutions and for this I was hoping to use a hall sensor. Also I will use two oxygen sensors that output a 5V analog signal, and finally some sort of current sensing device that measures the pulses to the spark plugs(two of these devices).
··I have no experience at all in the·programing of microcontrollers and am doing this as a hobby/ learning experience so there are no deadlines on production. So my main question is can the basic stamp support data acquisition between 5 sensor inputs? The spark plugs on my bike will fire around 124 times a second at max, and as of now I have no idea what sensor I should use to acquire those pulses so I don't know if the signal will be analog or digital.
··Any help you could give me would be greatly appreciated, but I'm really hoping you might inform me how to set up such a system. Will I need to chain several basic stamps together or can I get by with just one?
08-30-2008, 07:06 AM
Depending on the sample rate you want you might be pushing the stamp but since you have no experience in programming I would point you toward the propeller chip. It is chea.. less expensive and much more powerful than the stamp. Take a look www.parallax.com/tabid/407/Default.aspx (http://www.parallax.com/tabid/407/Default.aspx)
08-30-2008, 12:07 PM
As you, I have been working on a similar project, taking my sweet time at it, have made a hobby of it and am learning as I'm going.· Some of the things I've learned:
The Hall effect sensor that Parallax sells is a MLX90217 (MELEXIS, Co). It is meant to be used as a gear tooth sensor with a magnet attached to the back of the sensor, while the front faces an internal rotating toothed gear, probably inside a transmission case. Melexis has other hall sensors, one is the US4881, it is a couple of bucks less than the 90217 and is more appropriate for sensing a magnet attached to a rotating wheel, with the magnet passing close in front of the sensor. However, the 90217, can also be used, apparently, for that purpose. Here's the address of a guy that used the 90217 to measure (count) the rpm's of a motor flywheel and attached the output to a Parallax BS2, using the BS2's count command. (http://www.floridaame.org/Plans/Kamran%20Nili/rpm/tachometer.html ) It should be easy to apply this to a rotating drum.
There is such a thing as "clamp-on" ammeters. They measure current in a conductor by measuring the magnetic field surrounding the conductor while current flows through it. This type of ammeter uses a magnetic sensing loop that surrounds the conductor. Perhaps this type of sensing device could be used to detect pulses in spark plug wires, in fact I think I've seen something like this advertised on the internet, but I can't remember where.
On my Harley sportster, and other bikes, the electronic tachometer takes pulses from the electronic ignition and converts the pulses to RPM.· Basically from the low side of the ignition coil. But, you will need a voltage dividor to bring the voltages down to proper levels for the BS2, 5 VOLTS OR LESS. (the ignition gives off two pulses per RPM, and can be counted.· I have used the count command on my BS2 to count these pulses. The counts can be stored on the BS2's EEPROM)This address can help: http://www.nightrider.com/biketech/ (http://www.nightrider.com/biketech/)· He's the guy that gave the advice on the voltage dividor.·The harley's·speedometer uses a Hall effect sensor.
There are tachometer chips LM2907, LM2917 (frequency to voltage converters) that will convert pulses to a varying DC voltage, they were made for use·in tachometers.· Lots of info on the internet on them, just google LM2917.· You will find many circuits using those chips.· They convert to varying DC, which is analog.· If you want to convert from analog to digital, you can use an analog to digital convertor (ADC). I believe Parallax sells such a thing.·
There are other chips that you might find useful.· They are circuits that are already built and very inexpensive, and easier than trying to build them your self, and probably less expensive.· A very common one: 555 timer (Radio Shack, buck and a half or so) many uses for this, (multivibrators, Schmitt triggers)· Lm339 Quad comparators (Radio Shack, couple of bucks) could come in handy (Simple A to D converter :: Parallax forums ) using LM339.·· I tried it, it works.· The LM2917 tach. chip, I bought from DIGI KEY couple of bucks.· I have bought other things from DIGI KEY through the internet.· I told them that I am just a hobbiest and only need a few parts, their shipping charges are very reasonable.· DIGI KEY USA is in Minnesota, and can be found on the internet: google DIGI KEY.
I think I have uploaded the voltage dividor circuit that I got from http://www.nightrider.com/biketech/ (http://www.nightrider.com/biketech/), I hope it made it through.
I'll stop here
08-31-2008, 07:38 PM
Thank you very much franklin and po2le. I will take you advice and do a little more research. One thing I would like to say is that if I get this project off the ground I plan on making the whole thing open source so that anyone who wants a dyno cheaper than seven thousand dollars can do it themselves. Your input is greatly appreciated.
09-07-2008, 07:41 AM
·I think the info. po2le posted was very usefull. I have added some test programs that maybe of help to you.
09-08-2008, 01:16 PM
I recommend to use the Propeller, for this project. All the base code you need is already in the object exchange. Secondly, you need to stop thinking about RPS.·You need to·start recording·a revolution in micro seconds in order to maintain reliable data. Your variable (time_us) ·will be the basis of all your calculations. In order to record this data, you need edge detection from either the SX or the Propeller. But since you want 4 or 5 sensors too, the Propeller and Spin will be the easiest.
I would also recommend against the fuel air sensors, I think they would be a better choice for a brake dyno instead. I strongly recommend you learn to use Cylinder Head Temp thermocouples·instead and Exhaust Temp Thermocouples to verify detonation or preignition and to help fine tune your fuel mixture.
If you really want to use the fuel air sensors, create a data acquisition "street rig" and develop patterns to help you with CHT and EGT's. If you are racing, I may be able to help with some good #'s.
Also you need to make a good weather station, using the same Propeller. Be sure to use Absolute Barometric pressure (or a manual barometer corrected for gravity) , a·reliable temp and humidity sensors·and also be capable of calculating density altitude (if your planning on using a correction factor)
Post Edited (Capt. Quirk) : 9/8/2008 6:25:59 AM GMT
09-09-2008, 03:52 AM
· Why thermocouples instead of o2 sensors? I understand that the temperatures will vary between lean and rich mixtures, but with an o2 sensor I should get exact readings off center (or stoichiometric mixture).
· Yes I know to measure in milliseconds, however; I could have said .12 times a·ms or just approx 124 times a second. I stand by my decision.
· Also I had planned on building a weather station however I do not know the equations for SAE correction. Any help with those would be appreciated.
· Thank you for the programs.
09-09-2008, 11:38 AM
Thermocouples that mount under a spark plug are just plain easy. 300 degrees at the spark plug works very well universally on engines, 250 is a little rich and 330 is lean and mean.·Those measurements work at any rpm. Also·I have found CHT measurements to react very fast to small changes in fuel mixture·and be easier to interpret·than an EGT.
On Fuel Air sensors and Inertia Dyno's I am very opinionated, I don't like either·http://forums.parallax.com/forums/posticons/icon6.gif. I don't trust a device that's designed to be put in a specific location/distance from the piston and work at a specific temp. But, if I was able to verify·it's calibration, it would be different story.
The correction factor information I have from SAE (from 78') is outdated, I will check another source.
Post Edited (Capt. Quirk) : 9/9/2008 5:02:16 AM GMT
09-10-2008, 01:58 AM
What are you planning on using the dyno for?, if you don't mind.
09-10-2008, 08:14 AM
· Well a friend of mine got his wife a bike... I have one and he has one also, so does my roommate. We got to talking about modifications, all of us are college students and have plenty of free time when school is out but no money, so why not build a dyno (apply our expensive education) and see what cheap modifications can be done to our bikes and get a little extra power out of them. So we aren't doing this for production or any sort of monetary gains. However, in my research for this project I have found that most production dyno's, at least for motorcycles, are really over priced.
· I said all that to say this. I want to build it just to say I built it (it would really look good·in my major) and on the cheap. Also everything I do I want to document 100 percent so that anyone else can do it and add to it, making the knowledge even cheaper and easily accessible.
Post Edited (aerodim) : 9/10/2008 1:22:34 AM GMT
09-10-2008, 09:44 AM
I would recomend to save your time and money and just build a data aquisition system, use all the same formulaes and calculate horsepower while your riding·street bike. It's the same thing, just no dyno
If you really want a dyno, there are lots of Inertia Dyno How Too's already on the web.
Post Edited (Capt. Quirk) : 9/10/2008 2:50:59 AM GMT
09-11-2008, 06:12 AM
Thus far I have only found three homemade inertia dynos on the web and none of them are well documented. Maybe this is a fool’s errand.
09-11-2008, 11:58 AM
There·is a·how-to on building your drum style dyno(I believe the guy had a Suzuki in Australia or Europe), more dyno information from the karting guys (go-karts)(well documented, but not a drum), and the rc model boaters(also not a drum). There is also a well documented paper, possibly a thesis from a model boater too. Also more information on HP Here (http://wahiduddin.net/calc/cf.htm)
Here is another site (http://www.ajdesigner.com/phphorsepower/horsepower_equation_trap_speed_method_horsepower.p hp), that supports my idea that you don't need to build a dyno. Because an inertia dyno is no better than using your existing bike, within a set testing condition. A little massaging of those formulas and you·can plot an entire curve in 1 or all gears.
(in my opinion)
The only accurate information an inertia dyno provides is; the time (in seconds)·it takes to reach peak hp for a given load, not the measurement of Torque and HP.·It also does a good job of determining a shift point after peak HP. (for example)·If I built 2 pipes and both make the same HP, then I would use an inertia dyno to determine which pipe accelerates the fastest. (I don't trust test riders either)
This should be enough info to find it on your own.
Post Edited (Capt. Quirk) : 9/11/2008 5:34:01 AM GMT
09-12-2008, 02:42 AM
09-15-2008, 01:41 AM
I will add more as I find them
Post Edited (Capt. Quirk) : 9/14/2008 8:42:58 PM GMT
09-28-2008, 06:15 PM
· Sorry for the long delay in response time. Classes have started again and so my research has come to a close for now. I hope to make a decision on this during Christmas break and maybe complete the project next summer depending on what I finally decide to do. Again Capt. Quirk thank you very much for your help. I hope I haven’t·inconvenienced you any.