Solenoid Hit n Miss

[rudydubya]

Glad to hear the medication helped you, Chuck.  That's something I'll ask my doctor about next visit.

Following along and looking forward to seeing your latest effort running.  I love to watch those slow-running hit and missers, no matter how they're powered.

Regards,
Rudy

[zeeprogrammer]

That's excellent news Chuck. I'm very glad to hear it. 

[cfellows]

Here's a video of the engine running.  I have just one of the potentiometers hooked up, the one that controls the threshold rpm at which the engine fires.  After some deliberation, it doesn't really make much sense to hook up the other two controls.  I pretty much always want the solenoid firing for a full stroke.  Since the solenoid stroke is just 90 degrees, I calculate the duration of the power pulse to be 1/4 of the amount of time the crankshaft takes to make a full revolution.  That duration is going to vary with the threshold RPM and, since I can't measure angular position of the crankshaft, I have to use time.  The timing adjustment also seems a bit redundant since I think peak performance will always be when firing at TDC.  I may try to wind a solenoid with more amp turns to give the engine a bigger kick when it fires.  Don't want to make it too big, however, as it would ruin the aesthetics of the engine.

<a href="https://www.youtube.com/watch?v=cQvgFEiakmg" target="_blank">http://www.youtube.com/watch?v=cQvgFEiakmg</a>

[rudydubya]

Nice work Chuck.  Very innovative too.

Regards,
Rudy

[zeeprogrammer]

Very nifty Chuck. 

[Alan Haisley]

Here's a video of the engine running.  I have just one of the potentiometers hooked up, the one that controls the threshold rpm at which the engine fires.  After some deliberation, it doesn't really make much sense to hook up the other two controls.  I pretty much always want the solenoid firing for a full stroke.  Since the solenoid stroke is just 90 degrees, I calculate the duration of the power pulse to be 1/4 of the amount of time the crankshaft takes to make a full revolution.  That duration is going to vary with the threshold RPM and, since I can't measure angular position of the crankshaft, I have to use time.  The timing adjustment also seems a bit redundant since I think peak performance will always be when firing at TDC.  I may try to wind a solenoid with more amp turns to give the engine a bigger kick when it fires.  Don't want to make it too big, however, as it would ruin the aesthetics of the engine.

Chuck,


Maybe the timing adjustment should be a delay rather than an advance. You say the solenoid is fast; having it activate at TDC may lose some power. It might be interesting, anyway, and easy to do since it's there already. 


Alan

[PStechPaul]

Interesting, and a good runner! It seems that you must have a spring between the solenoid armature and the "piston" on the connecting rod. So the solenoid operates quickly to full end stroke and the maximum force will depend on the energy stored in the spring. It might work better to use a heavier spring so that the solenoid is slowed down by the inertia of the flywheel. In fact, you might not need a spring at all. To get smoother action, you could use PWM to apply more current at the beginning of the stroke where the force is lower than at the end of the stroke.

If the stroke of the connecting rod is longer than the stroke of the solenoid, you could use a mechanism that allows the connecting rod to freely travel beyond the solenoid stroke, and also allow the inertia to push the armature out past its usual limit. Then you could actuate the solenoid between, say, 45 degrees and 135 degrees where it will have the greatest mechanical advantage. The maximum speed will be limited to how quickly the armature can move, which will be a function of ampere-turns and the inertia of the moving parts. The current will rise at a rate determined by the voltage applied and the inductance, which will increase as the armature enters the coil. A solenoid will probably be able to actuate fully in about 20 mSec, which may be for 1/4 revolution, so maximum speed should be in the order of 1/(0.020*4) = 12.5/sec or 750 RPM.

If you have a scope, it would be interesting to see the current waveform, which should change with applied voltage and armature position. You can probably reduce the power and heating by cutting back the current at the end of stroke, at which point you only need enough to keep the spring compressed, or with a solid connection, it could be turned off completely, since the energy will have been transferred to the flywheel.

[Don1966]

Awesome Chuck!...

Don

[zeeprogrammer]

I find this interesting in part because I'm having to deal with a solenoid at work.
Can't have it on for too long else it heats up and melts the plastic housing.
Have to PWM down after activating it.

[PStechPaul]

A "quick and dirty" technique is to use a resistor and capacitor in parallel to drive the solenoid. Choose the resistor so that it provides just enough current to "hold" (as for a DC relay), and a capacitor with a time constant to the solenoid resistance long enough for it to pull in. PWM is better, but requires more components and "know-how".

[MJM460]

Hi Chuck,

Glad to hear you are feeling better.  Thanks for telling us about it, men are not generally good at looking after their health, and it does not serve us well.

Your solenoid engine experiments are most interesting.  I am another that would encourage you to wire in those extra potentiometers.  You have done most of the work by mounting them, and have the right soldering iron.

My thinking is as follows - while I am not sure of the form of a force vs armature position curve for a solenoid, or current vs position, a given force on the "piston" moves through most distance in any small time increment and at the highest speed when the crank is at 90 deg from tdc, and hence does most work and supplies most mechanical power at this angle.  Thus, the solenoid will potentially input most of the required power with a much shorter pulse centred on the 90 deg position, and potentially less energy drawn from the battery.  So it would be interesting to see if the engine runs just as well with both delayed energisation and a shortened pulse, saving electrical energy and solenoid heating.  I suspect your Arduino board also has a PWM function of you want to experiment in that direction.  Much more convenient than fixed components, since you already have the hardware.

I would not be worried about not having angular position, even very sophisticated industrial equipment uses an inferred angular position based on a one pulse per revolution signal.

When the energy (yours) returns sufficiently of course.  Priority is your health.  I am sure I can say we are all wishing you well and glad to hear of your progress.

MJM460