PDA

View Full Version : Truck Alternator to RPM



Sutton Murray
11-18-2006, 08:00 PM
I want to·connect my stamp to a truck alternator 24volts. I want to measure the RPM speed it turns at. I have been surfing the web to get ideas on how·this is done.·But I am probably looking at the wrong sites, i am at a·stage that i want to give up.

Can any one help me on how this is done?

Bruce Bates
11-18-2006, 08:17 PM
Is there just one belt driving the alternator, or is it one of those serpentine disasters?

Regards,

Bruce Bates

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
<!--StartFragment -->

Sutton Murray
11-18-2006, 08:46 PM
Well it was not really aimed at a specific alternator, but more in general. Those looked at by myself has one belt driven. Why, this question?

Hulk
11-18-2006, 11:00 PM
I haven't actually tried this circuit, but it's designed to extract and amplify the 3 phase ripple that comes out of the alternator (in addition to the DC).· I didn't design this circuit, I found it while surfing links originating from this board.
Hope it works for you.· If it does, please post your results.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔

Ken Gracey
11-18-2006, 11:18 PM
What about measuring the RPM from the device which is turning the alternator? If you know the pully sizes then you could probably measure engine RPM from a magneto pickup.

You could put a magnet on the shaft of the alternator, like this:
http://www.parallax.com/html_pages/robotics/machining/RPM_display.asp

Or if you use the magneto approach from the engine:
http://forums.parallax.com/showthread.php?p=612409

. . . thinking out loud, is this an internally regulated one-wire alternator, or is the field accessible outside of the alternator?·It may be possible to access the field·for yet another signal.

Ken Gracey·

Beau Schwabe (Parallax)
11-18-2006, 11:46 PM
Hulk,

That circuit is one that I posted. Here is the original link...

http://forums.parallax.com/showthread.php?p=566251

...5mV is very small. If you look further down there is another circuit version that
is sensitive at 600mV

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe (mailto:bschwabe@parallax.com)

IC Layout Engineer
Parallax, Inc.

Post Edited (Beau Schwabe (Parallax)) : 11/18/2006 4:52:00 PM GMT

Bruce Bates
11-19-2006, 12:42 AM
Hulk -

You could place a white stripe on the belt or on the back of the alternator pulley. Then just use an I/R retro-reflective optical transmitter/receiver. As I remember Sharp Electronics has some nice, small, sturdy units. If you need a link, or typical catalog number let me know.

This technique will work just as well for a gas or a diesel engine, and the use of I/R reduces the impact of any ambient, visible light. It should be relatively low cost as well. Just be careful of motor/alternator vibrations when you construct the physical mounting for it In other words, don't "stem" mount it, use a multi-point (say "V") mounting. Otherwise you may end up with a vibrating reed type device http://forums.parallax.com/images/smilies/smile.gif

Once you have the signal from the optical gate, run it into a Stamp pin, and use the COUNT command to develop the RPM. Nothing could be simpler. If you need to clean up the signal (which I rather doubt) there are simple logic gates which will do that for you.

Regards,

Bruce Bates

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
<!--StartFragment -->

Hulk
11-19-2006, 12:58 AM
I didn't start this topic so I can't speak for Mr Murray, but I appreciate the feedback as this is coming back as an urgent need in the near future. Thanks gents!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔


·

Sutton Murray
11-19-2006, 06:32 PM
Through the reply to my question for help, i found it clear that a lot of discussions about this happened in the past. I have learned a lot from all of your replies. A lot of debating in the past: with a lot of amazing ideas.

Hulk Thank you for the circuit. I will be trying it out.
Beau understanding the circuit belongs to you, many thanks for the posting of it.
Beau from your previous forum discussions your last posted circuit the presumed 600mv circuit, i will be using this to do my tests. I do not at this stage understand which one of your 2x circuits will be better for my use, but i will be definitely be looking in to it.



Ken I am a lot on the parrallax web site. Never saw the CNC link as given by you. Just shows you how bad i am on look through the web sites. Your second power plant is just as Awesome. Compliments: from the photos and videos i will say perfectly build.
Thoughts that past my mind, i wondered how your power plant project will do as a 120 volt AC plant, see the link below and you will understand where i got the thoughts from. "Alternator secrets" by an unknown author
http://www.1stconnect.com/anozira/SiteTops/energy/Alternator/alternator.htm

The alternators will be mostly internally regulated, as to Ken loud think out. As my aim of use of this device will be fitted to many and not used on one application. So it will be a lot more suitable not to do any modifications to any of the Diesel engines of concern. I wanted to move away from the magneto pickups as all the engines do not have this fitted. Fitting the pickups will enlarge the cost of the unit. Low cost to complete build of this device is as important to me as to its use of application and reliability.
Bruce thanks for your idea. I will not go with this as to the planned applications are for underground mining machines. Dust is of the utmost extreme in these environments. I believe the I/R reflector will not even last one operational shift before the dirt will stop things from working.



I will be back after tests. Either asking more help or if i am successful with the posting of the complete testing to what i am aiming for.

stamptrol
11-19-2006, 07:35 PM
Sutton,

A couple of other things I've seen over the years.

For best results, you really have to get your signal before the bridge rectifier in the alternator. That way you'll see a definite AC waveform which is easy to measure. But, it means you will have to have the alternator modified with the AC tap. However, many industrial alternators use an external diode bridge so its easy to get at the AC connections.

Once you've got the AC signal, you can use a chip like the 75176 RS-485 receiver/driver to give a nice clean square wave for the Stamp to process. It can handle bi-polar input signals quite nicely without any external components.

An alternative is to use an inductive proximity switch to see some rotating part of the engine. It is immune to dust and dirt and will also give a good signal for the stamp to work with. Just check the max frequency number on the prox switch because some are low speed only ( Telemechanique are the ones I use).

Cheers

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tom Sisk

http://www.siskconsult.com
·

TC
11-20-2006, 08:39 AM
I have to put my two cents in.

As you might tell I am in the automotive audio/video industry, back in the early 90's JVC made a radio that will turn up the volume as the RPM's increased. and now since 2005 pioneer has made moble video units that read the RPM with no connection to any RPM source. and meny OEM radios will turn up the volume with engine RPM.

they work by measureing the noise that comes from the alternator, now how they did that, I don't know!!!

but maybe someone would be able to figure it out.

I have to find meny ways to interface the OEM world with the aftermarket world. most 24 volt trucks (that I have delt with) have a tach wire at the gauge cluster, the hard part is finding it.

take a DVM set to AC volts, one lead on ground, the other lead·start testing all wires on gauge cluster with engine running. when you find one that changes voltage when you press the gas pedal, you found your tach wire. now I do not know how you can use that but someone here sould be able to help.

TC

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
We all make mistakes when we are young………That’s why paste is edible!

bcw1011
11-21-2006, 02:21 PM
heres a bit more info....


most spark ingnition engines generate more electrical noise than the diesel conterparts do. (the ignition system is to blame) so the two ways you can 'detect' the engines rpm is by filtering out the electrical noise from the ignition system or to amplify and filter out the AC ripple from a alternator the alternator's diode trio converts the AC voltage into a series of spikes.... in other words the signal goes from /\/\/\/\/\/\ to ^^^^^^^^ the vehicles battery along with other components will help to smooth out the ac ripple making this effect less noticable AC ripple can also cause noise on a radio. you can see this waveform along with all the other electrical noise by using a oscilloscope.


also ome types of alternators have the wire to provide a tach output. a field connection should also work. it sounds like your using a diesel engine. some types of diesel engines also have a provision for a magnetic/inductive pickup using the starter's ring gear as a encoder wheel. some other things to think about.....if your engine is ECM controlled you might want to check into how that operates, you might be able to utilize a injector output or other means (with the proper circuitry). or if your engine is a mechanical injection you can make a external sensor to pickup the vibration from the injector lines (same way some timing light adaptors work)


I work with AC generators for my day job so seeing 24volt engines is almost more common than 12v what kind of engine & alternator are you working with?
-Brad

Sutton Murray
11-21-2006, 05:34 PM
Thanks every one for the help. To some degree all of you do make sense. But hey I am no electronics expert and I want to succeed in building this.
What about Beau schematics will this do the Job? I have attached 2x picture (alternators) used on the Deutz Diesel engines. Although these are not the only alternator types used, but at leased a starting point. The alternators used are mostly Bosch. Look at the attached picture “Alternator 1”, can I attached Beau schematics directly to the alternator?

Beau Schwabe (Parallax)
11-21-2006, 11:04 PM
Sutton Murray,

Can you scope the output of the Alternator to see how much ripple you are going to have? ...Note: I would probably lean more towards the 5mV circuit.
Depending on where you tap into the automotive circuit i.e. closer to the Battery vs. closer to the Alternator will also be a factor as to how much "ripple"
will be noticed by a scope or any other type of circuit.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe (mailto:bschwabe@parallax.com)

IC Layout Engineer
Parallax, Inc.

Capt. Quirk
11-22-2006, 11:51 AM
Wouldn't be easier to buy a $30.00 "Tiny Tach". they wrap around a plug wire,·they measure rpm and doubles as an hr meter. Otherwise, an OEM·BS2 stamp or an SX (to keep the cost down)·and the melexis hall sensor using the pulsin command should be a simple install.

Sutton Murray
11-22-2006, 05:38 PM
Beau, i will do some test with a scope. I have a parallax scope; this should do the job right?

Bill if i understand you correct you referring to the plug wire as used on a gas engine. These are diesel engines! As previously mentioned i am trying not to fit any speed sensors. The unit will be fitted to multiple engines. All the engines are not exactly the same to each other. To exclude the sensors, this will also make installation easier.

Benoit
11-22-2006, 11:08 PM
Sutton, you want to be careful with your o-scope. If you have the Parallax USB o-scope, it's limited to 20vpp on probes 1 and 2. You mentioned that the alternator is 24v, so I think you should add a resistors in series before you hook up the o-scope to the alternator.

-Ben

Post Edited (couch-pilot) : 11/22/2006 4:12:43 PM GMT

Beau Schwabe (Parallax)
11-23-2006, 12:34 AM
Sutton Murray,
Couch-pilot (aka Ben) is correct, the input to the USB scope is limited to 20Vpp.· However in order to increase the amount of allowed input voltage, it's a little more involved than placing a series resistor on the scope probe.
The first thing you need to consider is the input impedance of the scope, which happens to be 1Meg.· Since the scope is calibrated and matched to this impedance, we really should maintain this value.·· If we place an external
resistor of any value in series with the probe it does form a voltage divider, but it also increases the impedance by the resistor value we put in series plus the original impedance.· That said, you could place a 1 Meg resistor
in series to effectively form a X2 probe, but now your impedance to the scope is 2Meg.· Something like this....


R1 R2
Probe <---/\/\---o---/\/\----> GND
|
o-----------> Scope Probe

R1 = R2 = 1Meg
R2 = Scope impedance
R1 = External Resistor


...In order to bring the impedance back down to 1Meg, and still have a X2 probe, you need to create a second voltage divider in parallel with the first.· Something like this....



R1 R2
o---/\/\---o---/\/\----o
| | |
Probe <---o | o----> GND
| R3 | R4 |
o---/\/\---o---/\/\----o
|
o-----------> Scope Probe

R1 = R2 = R3 = R4 = 1Meg
R2 = Scope impedance
R1,R3,R4 = External resistors


The attached image is basically a X2 scope, that maintains the scope impedance·of 1 Meg.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe (mailto:bschwabe@parallax.com)

IC Layout Engineer
Parallax, Inc.

Sutton Murray
11-23-2006, 03:39 AM
Hehe
·
I have read your warnings·now only after the tests.
·
Now that it was mentioned to me, now only i·will see the "20Vpp max" stamped on the scope.
·
What happens if it wasn't done as specified by you?
·
See attached results

Beau Schwabe (Parallax)
11-23-2006, 04:36 AM
Tell us what you have connected. Is this directly off the alternator? IS the battery connected to the alternator to help smooth out the ripple? Are your test leads closer to the battery, or closer to the Alternator?

"What happens if it wasn't done as specified by you?"

You would see clipping... the impedance is high enough (1Meg) that the amount of current for voltages over 20 is going to be small unless (I = V / R) at 20V, the current is only 20uA

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe (mailto:bschwabe@parallax.com)

IC Layout Engineer
Parallax, Inc.

Sutton Murray
11-23-2006, 05:16 AM
My connection; the "alternator 1" was used for the test. This was directly done off the alternator. No battery was connected to the alternator. Test leads were fitted close to the Alternator. As in the picture "Alternator 1" the test lead was grounded and connected to terminal marked "W".

With the "W" terminal being the AC output. Will a connected battery to the DC terminals have any affect on the AC output terminal?

Beau Schwabe (Parallax)
11-23-2006, 05:22 AM
The Battery Is going to act like a large capacitor and "smooth" out the ripples produced by the alternator, so yes your readings should look very different.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe (mailto:bschwabe@parallax.com)

IC Layout Engineer
Parallax, Inc.

Clay L
11-23-2006, 07:23 AM
Just curious, why are you wanting to build the tachs rather than buy commercial ones made to connect to a tach tap on the alternator?

Clay

Post Edited (Clay L) : 11/23/2006 12:29:19 AM GMT

Randy
11-23-2006, 08:35 AM
The enviroment might be a little tough for such a sensor and I have not looked to see what is sutable but has anyone thought of listening to the exhaust? If you catch the explosions leaving the engine, with some simple math you could deduct rpm. Depending on location, manifold, one cylinder, several cylinders, tail pipe, single or dual, etc a programing option would alter the math to match the aplication. You could possibly mount a sensor with a·hose clamp and it wouldn't care if it was gas, diesel, two stroke, four stroke as long as it goes bang.

Randy

Sutton Murray
11-23-2006, 02:28 PM
Clay-It is all about cost.

Randy- it sounds like a good idea. Do you have any idea on a circuit, sensor to be used? I also can't at this stage imagine the code to be written. Have you seen this being used some where?

Beau- I want to do my next tests. This time I also want to attach the stamp with code to read the ripple. I will modify the scope as described and attaché this as well. Obviously I will have a battery connected. What I am still a bit unclear about is the circuit between the stamp and the alternator. Your circuit posted by Hulk, I will use not that one but your second improved circuit as found in a previous forum of yours. Am I correct if I say that this circuit will be needed due to the ripple smoothing after the battery is connected? But say if no battery was used that I can exclude your circuit and connect the alternator direct to one of the stamps pins?

Clay L
11-23-2006, 10:39 PM
A little bit of history – many years ago I designed electronic tachs, speedos, and other instrumentation for heavy equipment.
I would not connect the Stamp pin directly to the alternator field coil tap. Some pretty healthy voltage transients can be seen at that point. I would use some sort of conditioning circuit between the tap and the Stamp pin.·
Its been a long time since I did the alternator tach design, but as I recall I used a low pass filter consisting of an input resistor connected to·a capacitor with the other side of the cap connected to ground. The connection between the resistor and the cap was the input to my circuit. I can't recall the values, but probably something like 10,000 ohms for the resistor and maybe .1 mfd for the cap.
I don't know what your budget is, but there are two other approaches you might want to consider.
Magnetic sensors are pretty cheap -·a coil of wire around a magnet potted in epoxy - and many diesel engines used to have a threaded hole in the flywheel housing for such a sensor. The output would be a sine wave with one cycle per tooth on the flywheel.
Diesel engines used to have a threaded outlet for mechanical tachs.· I designed a sender that fit that outlet and·output a sine wave with two cycles per rotation of the sender. The part number is something like a 201A sender generator and they are still manufactured by Dixson Inc (division of Datcon) in Grand Junction CO.
·
One other note, with the circuit·resources I had the the time, I found it best to use the period of the input rather than the frequency of the input as the basis for my rpm measurement. I haven't thought the problem through using a Stamp to count the input frequency and convert that to rpm, but if you run into problems you might want to consider that approach.
Clay

Randy
11-24-2006, 10:36 AM
Sutton,

I gave no thought to a particular sensor for listening to the exhaust. You said you wanted this to be useable on several engines and types and I thought only as far as the one thing that is common to all internal combustion engines is that they have strong exhaust pulses. Lawn mower, airplane, boat or car - bang, bang, bang divided by the number of cylinders times two for a four stroke and you have rpm. There are lots of transducers that work off of sound or vibration the first trick is to find one that will survive the heat and mounting method. It would have to have the resolution to be able seperate the exhaust from all the other noise an engine makes. I would guess that on the exhaust pipe there is a substantial differance between exhaust and other noises but on the manifold I don't know how much of the other noises might be a problem. You might try using a mechanics stethoscope to get a rough idea.

Some engines use a 'knock sensor' to adjust the ignition timing when it 'hears' preignition. As an OEM part I am sure it is not cheap but a search might give you an idea of how they work and what they are made of. Might give you some mounting or location ideas too.

As to the circuit there are a number of ways to count frequency or time intervals. I am only vaguely familiar with them. Code? What are you doing with the information? If you are just displaying rpm you might not even need a Stamp. Just an interval counter, display driver and the display.

Just some ideas for you to ponder. Sorry I couldn't give you any specifics but this is just a hobby for me. In fact I am about to post a question I have regarding ohm's law, beginers stuff.

Randy

PJ Allen
11-24-2006, 10:26 PM
Sutton Murray said...
I want to·connect my stamp to a truck alternator 24volts. I want to measure the RPM speed it turns at.
He's not after the engine RPMs, but the alternator's.
I'd go with a piece of foil or paint dot·on the pulley and rig up a LED-&-photodiode: LED light, reflected off the foil, into the photodiode = event.
I'd be surprised (but not shocked) to learn that there are off-the-shelf, ready-made "sensors" [e.g. the Parallax Truck Alternator RPM Sensor AppMod plug-in] for something like this.· You're going to have to be, get ready for it,·an·Experimenter.

Post Edited (PJ Allen) : 11/24/2006 3:35:02 PM GMT

Beau Schwabe (Parallax)
11-25-2006, 01:21 PM
Sutton,

You said...
My connection; the "alternator 1" was used for the test. This was directly done off the alternator. No battery was connected to the alternator. Test leads were fitted close to the Alternator. As in the picture "Alternator 1" the test lead was grounded and connected to terminal marked "W".

With the "W" terminal being the AC output. Will a connected battery to the DC terminals have any affect on the AC output terminal?
Took awhile for this to sink in.... "No battery was connected to the alternator."· ....huh? I thought alternators required an external power source for the "excite" coil... and they
didn't contain any permanent magnets.· This could explain the no more than 3V Pk-Pk on your scope outputs, if the armature was slightly magnetized.

If you are not connecting power to the alternator or using it to generate, then I'm a little confused as to why you would have one in the first place.

Can you re-state your goal here?

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe (mailto:bschwabe@parallax.com)

IC Layout Engineer
Parallax, Inc.

Rosco
08-08-2012, 01:15 AM
Hallo,

I know that this topic is pretty old but I have a question about the Optascope 81M as well.
It is about the FFT-function which I can't find in version 5.1.1.
Please click this link to find my question in more detail.

http://forums.parallax.com/showthread.php?141713-Optascope-81M

(http://forums.parallax.com/showthread.php?141713-Optascope-81M)The given examples about the probe-divider to double the scope-limits was very helpfull because that was another thing I was looking for.
Tnx

########

I think I found my answer in the same link I showed above.