A few weeks ago I had a wonderful opportunity to meet with Ray Y., a mobile tech in Toronto and a regular contributor here at Diag . Net. Ray showed me notes of his recent diagnostic cases, and one of them included a somewhat unusual waveform for a transmission solenoid. The more you look at it, the more interesting it seems.
Here it is (EDIT: better resolution file):
An amp-clamp shows 3.1A (EDIT:corrected from 3.5A) current until the solenoid is turned OFF.
And here is the quiz. Remember, if an answer looks obvious, it is probably not correct :)
- What is the reason (or reasons) the TCM is using PWM to control the solenoid?
- What is the purpose (or purposes) of the inductive kick?
- How is it possible there is no inductive kick during the OFF period of the PWM? Is there a reason for such circuit design?
If you need any clarifications, please ask. Enjoy the puzzle and thanks to Ray for providing the waveform.
Well, here goes nothing.... - The TCM is using PWM to control the solenoid for more precise control of the solenoid. Also allows controlled slippage (for instance a clutch apply solenoid). - It seems counter-intuitive to me, as inductive kick is avoided in general. but perhaps it is to prevent false signal feed back to the TCM. - Meaning, that if a large fly-back diode was to used to
- Energy efficiency while accommodating the fast response of a low-resistance winding. - No purpose, per se, but result of design for other intents. - The magnetic field doesn't collapse. I am not sure what portion of the design you referred to in the second question.
Hi Dimitriy. Interesting. Just looking at at the numbers using 15 volts and 3.5 amps, Ohm's Law calculation results in a solenoid resistance of just shy of 4.3 Ω. That would be the reason for the high frequency shown in your "postage stamp" sized image without anything other than voltages displayed. The image isn't clear or large enough to see the duty cycle or other details. Solenoids
I'm glad Martin mentioned the image size. When I click on it it barely gets bigger. I can't see the amperage waveform. I have always believed that they went to PWM so they could use smaller (cheaper) wires, and smaller (cheaper) solenoids. I have talked with Ray Y several times and he seems like a great guy. I'm sure he enjoyed the visit!. Toronto is far too far for me. Not to mention COLD!
- PWM is used to control the solenoid more accurately; usually used for precise control of hydraulics, pneumatics, etc., for pressure control. - The last time I saw a use for an inductive kick like this, it was for actuation verification. The pintle moves enough, in combination with the field collapse, that a voltage is induced, telling the PCM that the pintle moved. - This is a guess, but I
Here's my guesses; 1. Limit current, finer control. 2. Some have no purpose and have to be suppressed. Others may be monitored for circuit integrity. Some may be attenuated by an avalanche diode. 3. Magnetic field does't have time to collapse during off period.
- to control position of spool valve (oil pressure control) - inductive kick can be used to verify circuit integrity, if monitored by ECU. It is due to the rapid collapse of a magnetic field and can infer health of the solenoid. - there is not enough time for the magnetic field to collapse before the circuit is energized once again
Dmitriy: Thanks for posting. Your questions are very good but, the circuit description and component information is incomplete. Some of others that have posted (general) answers to your questions but, to provide specific answers, I would need the following: - Frequency of the PWM signal. - Inductance value of the solenoid - Resistance value of the solenoid - Type of solenoid
UPDATE: Clarifications. Thank you for all the responses, I will try to add clarifications first: - My scanner default settings were, indeed, inadequate. I've tried to use the highest resolution, and the result is attached to the original post. But be careful with the current waveform -- it is a sum of several solenoid currents. There is also another scan showing what happens when the solenoid
It appears that the inductive kick occurs during the same time period as the "off" times. I'll stick with my original thoughts; although the capture does disprove, in my mind, the lack of full saturation of the solenoid windings. I'm here to learn, so I look forward to the answer.
It looks like each pulse period is about 1.7 ms. That seems more than enough time for a voltage kick. If you think about an ignition system, the firing kv spike is measured in microseconds. Therefore, there must be something more going on in the circuit. The spike must be being controlled in some manner by the circuitry. That's my theory anyway.
In this capture, the scanner was off and for some reason, the PCM decided to start cycling the LC sol, on and off, in Park, for 5 or 10 minutes, until I turned the key off. The PCM never cycled the other solenoids like this. It looks like the PCM is testing the LC sol. CH A is the PCM cycling the LC sol CH B and CH C are 2 other sol voltages CH D is the amperage on the sol pack feed wire
Hi Ray, thanks for sharing yet another capture. This one reveals solenoid resistance very nicely. This is Key-On-Engine-Off capture, so we have to be careful with the voltages. After zooming in (see attached), there are - Supply voltage is 11.6V (Ch B and C), visible on solenoids that are OFF. - Ground voltage is 0.7V (Ch A, cursor 2). - Pushing current of 6.3A (Ch D, cursor 1) through
Alright, I think we have enough info to address the first two questions. 1. PWM is indeed used for fine control of the force exerted upon the pintle, and pretty much everyone mentioned this. Marlin also mentions energy efficiency, and, yes, PWM is used for ON/OFF solenoids for that purpose as well. However, here using PWM is simply vital. Ed mentions current limiting, and Martin notes high
Dmitiriy, Do you think the voltage spikes are being suppressed until the PCM wants to test the solenoid?
Yeah, I think the PCM can control what kind of flyback-supressing circuit the solenoid will see. I will try to provide justification for this in the next update.
In my last capture, the scanner is off and the PCM decided to cycle the LC solenoid on and off, for at least 10 minutes, until I turned the key off. The PCM did not cycle any of the other solenoids. It would be good to know how to get the PCM to cycle the solenoids on demand, without the use of the scanner.
Was the code you have mentioned, P075A, active at the time? If yes, may be this is what explains the cycling of the LC solenoid... and if it is possible to set DTC for another solenoid, the computer will start cycling that one instead? (just thinking out loud, obviously)
Hello Dmitriy and All: There are number of items on this circuit that are undefined so, I'm going to make some assumptions (not always good) as part of my response Assumption #1: The transmission solenoid uses one spring to close the valve at rest. It could use 2 springs to center the valve (with the valve controlled in two directions) but, unlikely. Assumption #2: The solenoid is a
Mark, my mistake, if I knew better I would have specified that this quiz is geared towards diagnosticians/technicians. I will try to explain how. A diagnostician gets to the car, identifies a suspect component, may be finds some service info about it (often quite incomplete), collects a waveform. Now the diagnostician's quiz begins -- is it a good or bad component waveform, why does it look
Valid comments Dmitriy. As a long-time dealership flat rate technician, repair orders received were often worded in this way: "Power door lock doesn't work. Check and report". Service advisors on commission themselves, often do not obtain enough information about the customer concern themselves or perform a quick preliminary verification. That leaves the technician playing detective in the
Don't have time to read through all of the messages. This response is just general information. I apologize that I did not read not read enough to know what circuit you are connected to still I hope this contributing is helpful. "return the pintle back into the initial position fast and reliably". I believe there is a more accurate way to say it - instead of fast and reliably perhaps simply