Newbie has a question

Fatfenders

Hello,
I just recieved my BASIC Stamp 2 kit today and have spent the morning reading the manuals and inspecting the kit. I purchased this kit for a very specific application and am hoping for a little guidance here. I have built a 1936 Chevrolet Coupe street rod powered by a '95 Pontiac 3.8 Supercharged 3800 EFI V6 coupled to an (electronic) 4L60e 4 spd (overdrive) automatic transmission. I purchased and installed an aftermarket transmission controller (TCU) which has worked resonably well up until recently. But after only 16k miles·the logic to lock up the torque converter is failing in that lockup is random or non-existent. For those unfamiliar, transmissions like the 4L60e have a feature that locks up the torque converter at a specific speed if certain conditions are met (i.e. transmission·fluid is up to operating temperature,·actual 3rd or 4th gear). Without this lockup the fluid can reach much higher temperatures, reducing the life of the transmission. TCC lockup is achieved by simply closing ground to a pin on the transmission.
Now about my 'level of expertise'.·There is not much about automotive 12 volt stuff·don't understand and can't do, but when it comes to electronics, I have enough conceptual understanding·to build, for example, the wiring harness's for·my street rod·but I have precious little actual understanding of electronics.
OK, a new controller is over 800 bucks, mine is working OK except for the TCC (torque Converter control) so I thought I would try'n cobble in the stamp 2 to control just the TCC stuff and maybe learn something along the way.
Now, that was a long winded introduction to my·first actual question.·When I talked with Tech Support they said the stamp 2 could read 'pulses'. The 4L60e VSS (Vehicle Speed Sensor) has a reluctor on the tailshaft (see attachment) that shows '4000 pulses per mile'. I know the VSS is working properly since other systems that depend on it (i.e. the speedomenter) are operating normally. After reading about the Stamp 2 'PULSIN' command I·realized immediatly I need help! low pulse? high pulse? 0-1 transition? Oh no, what have I gotten myself into???

Larry~

I have built speedometers for gm's reading these pulses with the count command is no problem.

a few things come to mind is the brake electrical ok if you touch the brake· while the
converter is locked up it drops out check this line, also the tow/hall input will do the same thing.

Larry~

Ok I re-read your post, the output of the tailshaft is AC sine wave
and the voltage was higher than 5 volts if I remember correctly.

so you will need to condition the the output with a diode and clip the voltage
at somewhere arround 4.5 to 4.8 volts. then use pulsin or count commands to
read and turn on/off the overdrive.

You may need some engine rpm info also?

don't worry there's lots of good help here, it sounds very doable to me!

Pat Alonzi

Sounds do-able as well. You could use an OP-AMP to handle your signal conditioning once you place a diode to rectify your signal if it is indeed an AC wave. An oscilloscope could be really helpful but not incredibly necessary. The hard part is testing and driving at the same time. I used a Basic stamp to read speed input on my gocart (I built a Hall Effect circuit similar to what Parallax used in their example on the CNC milling machine) hardest part was testing the code while the go kart was jacked up on stands in the garage.

Fatfenders

Thanks for the replies but I don't actually know enough to be able to use your suggestions. "Condition the sine wave?' 'Clip the voltage?'. Use an OP-AMP?. I know enough to follow diagrams and read a VOM but not enough to translate these statements.
Another bit of information... This is from the Pontiac shop manual:

CIRCUIT OPERERATION
The vehicle speed sensor generates a signal that is proportional to vehicle speed. The signal produced is a sine wave, which the PCM processes into a square wave to supply inputs to the speedometer, the multi-function chime control module, computer command ride module and heads up display.
The PCM takes the voltage pulses from the vehicle speed sensor, through the PPL and YEL wires, and uses them to close a solid-state output switch. The output terminal is switched to ground at a rate that is proportional to vehicle speed.

LSB

Doesn't sound to me like you're ready for op=amp, AC, and diodes. In the off chance I'm right, lets look a little at what your trying to do. Keep in mind that I know nearly (not quite, but almost, as much about cars as you do about electronics). If you think (as I do) of electricity flowing in wires as water in pipes, then that will get you all of the way in understanding a brake light, right? Electrons, like water, flow from a reservoir (battery) through the conduit (invisibly) to where a valve (switch) turns them on or off. "On" the bulb glows and "Off" the bulb is dark. We can "T" electricity like water, run it through various mystical control apparatus, and general understand what is happening by relating flow to amperage and volume to voltage, right?

Electronics, to continue our representation, adds another dimension—signal. If you think of this signal as patterns of current in the water (like rings in a pond), it is no more correct, but useful in some aspects. The Stamp cannot handle more than 5 volts or more than a fraction of an amp, but it can read and create ripples and currents in this tiny flow of electricity that other electronic devices understand. Typically, these devices fall into two categories: sensors that collect data and actuators that respond to signals output by the Stamp.

To bring this all home, if I understand correctly, you want a sensor to monitor engine RPM and signal the Stamp—which will interpret, evaluate, and respond by outputting a different signal to control the torque converter lock on the vehicle, this based on the logic that you program the Stamp to apply, correct?

Diodes, op-amps, hall sensors, and high or low are all terms specific to the control of this signal—either to condition the 12V common to your vehicle not to exceed the capability of the Stamp on the input side, and to boost the tiny output of the Stamp to a level capable of effecting your transmission on the other.

If, as L_Gaminde asserts, the signal coming from the sensor is AC and higher than 5V, then a diode is just the device you need. Back to our rings in the pond example. A diode clips the “rings” so that only the tiny peaks slide through—forming a row of “blips” similar to the electric pulses that cause spark plugs to fire (but MUCH smaller)—perfect for the Stamp to count with—you may have guessed, the COUNT command. Based on the COUNT, well make decisions using IF and THEN and finally, use HIGH and LOW to make output signals to tell the transmission what it should do (lock or unlock).

My example is brief (I find typing tedious) and vastly over simplified, but I believe it will give you some context to study the many manuals available in the download section (“What is a Microcontroller” comes to mind) and begin searching Google.

Oh, by the way, computerized dashboards, keyless entry, remote ignition, and all manor of neat gadgetry is ALL based on the same concepts you’ll learn while fabricating your transmission lockout.

Fatfenders

I found some more information in another manual this evening... see attachment.
It states, in part, "... the DC signal is then sent to the PCM and interpreted as vehicle speed."
Does this mean that the signal out of the VSS is actually the 5v DC signal and that it could be be inputed directly to the stamp?

LSB

According to your documentation, yes, the conditioned signal is 5V, but the signal is AC to the DRAC. I don't know where the DRAC is, or if the signal is accessible after that; it may be necessary tap in to the signal prior and condition if for yourself. There’s more of a leaning curve in this home rolled approach, but the parts will probably cost less than $5.

Try tapping into the wires involved and see what your VOM tells you—see if you can find that conditioned signal. A 12V reading is not what you want, but anything between 1V and 5V has possibilities (a VOM is probably too slow to read individual pulses, so will average the voltage as in: if the pulses are “HIGH” (5V) ½ the time and “LOW” ½ the time (0V), the VOM will read 2.5 volts—read up on PWM for more info.

Once you have that (if you can find it at all) then the next step becomes clearer. Also, can you say for sure you know which wire carries the unconditioned signal? Perhaps a dealership or fellow motorhead has some insight or equipment that will help nail these things down.

Larry~

don't think so I think they condition the signal in the DRAC whatever and wherever that is. (digital ratio adapter converter)

I thought about this last night and you should take the signal out of the pcm or DRAC, this is a 5 volt square wave output and I think its·a grey or purple·wire out of the PCM to the speedo, gee its only been 10 years since I did this on an LT1, in a 56 chevy.

Fatfenders

L_Gaminde,

Hmmm, you say use the output from the PCM, but the manual clearly states that that pin is a 4000 ppm signal that is 'switched to ground at a rate that is proprtional to vehicle speed."

Doesn't the 'COUNT' command require a 0-1-0 type of input and not just a switch that closes to ground?

Also, I hooked up the + lead on my VOM to the (YEL) wire from the VSS and ground to the other and took it for a spin this AM.

Here is the result.

Key on or engine idle· .61v

10 mph 6.8v

15 mph 7.9v

20 mph 8.1v

30 mph 8.6v

above 30 ( to 55 mph)·there was very little change, hangs around 9.1 - 9.2v

Also, I failed to mention that when I built the car I used 'Classic Instruments'. The electronic speedometer came with a 'Dakota Digital' SGI-5 interface. This box takes the signal from the VSS and converts it to 5 other values. I run the Speedometer, Multi function chime control, and electronic cruise control off of it. Those systems all work fine. The way I have it configured the outputs are:

·128kAC, 128koc, 8000AC, 4000oc, and 2000oc

Larry~

Hmmm, you say use the output from the PCM, but the manual clearly states that that pin is a 4000 ppm signal that is 'switched to ground at a rate that is proprtional to vehicle speed."

Yep that sounds correct 4000 pulses per mile. switches to ground is correct so it goes high 5v then low 0 volt then high 5 volt this is the square wave that you want to see using pulsin or count commands.



·128kAC, 128koc, 8000AC, 4000oc, and 2000oc

what are these! pulses out what is oc and AC. just program your bs2 and hook to 4000oc and read speed.

it would be nice if you got a serial LCD three wire hookup then you could display speed while you test.

Its time to start programming so get after it.

Fatfenders

OK, I finally got it. Rather than use the VOM I cobbled in an LED·to the SGI-5 4000oc output. Waa-Laa! At very low speed·I could 'see' the pulses.·Hopefully this will be my last post for a while as I think I have enough understanding/confidence to get this thing working. Thanks for the help.

LSB

Ahh... Just a last few word of warning. Be careful that the 'Stamp never sees more than 5V or 20ma and, because vehicle power is often very "dirty" with spikes and noise, filter your power carefully.

Fatfenders

OK, 'progress', of sorts.
Got everything hooked up and working in the car using a 9v regulator and the 4000ppm output from the Dakota Digital SGI-5. Things seem to work about right·but with one problem and I expect the answer to be the 'noise' issue.
I found that if I 'COUNT' for 150ms, the count·is apx. 10X vehicle speed. Easy to work with. However, very randomly I will get non-zero counts·sitting still.·Its too random to really pin down but for instance, one time as I sat idling it was consistently producing >0 counts. I hooked up an LED and found the SGI-5 was in the 'low' part of the cycle. Then I hooked up my VOM and found that the SGI-5 output varied as high as .0020v.·I guess the stamp was counting the variance from 0 to .0020 as a cycle. But then later I could not duplicate it. Another time it was counting at idle I·disconnected my ground (so that was grounding only through the USB port > laptop > serial port > chassis ground and the counts went to zero, hooked my ground back up, >0 counts. Tried it again later, no affect.

So anyway, about this 'spike and noise' and 'filter' stuff... how do I translate your caveat into real stuff?

Thanks again,
Dave

bennettdan

Fatfenders,
I like the choice of motor in your ride I have the same type in my 96' Riviera, here is a websight that explains how to condition your input signal.
·www.ibiblio.org/obp/electricCircuits/Exper/EXP_6.html#xtocid256531
Look at the voltage comparator circuit what is does is allow you to set a voltage threshold your sensor has to reach before it sends out a high "on" state. what you can do is use a simple POT as a voltage divider and set the threshold say to 4v and when the other side of the op-amp sees 4v then it will output what voltage you set it with say 5v with a zener diode. I will see if I can draw a schematic if you need one. In the example in the websight lights an LED you can run this into your stamp by either make the voltage stay at 5v with a zener diode or us an optocoupler to keep the voltage safe. Here is a websight on the zener diode as a voltage regulator.
http://hyperphysics.phy-astr.gsu.edu/hbase/electronic/zenereg.html
Also here is a circuit that i drew up that should work just use a voltmeter and chaeck it out before you hook it to your stamp.
Good Luck

Post Edited (bennettdan) : 5/17/2007 6:26:47 AM GMT