How would you power this solenoid.
T Chap
Posts: 4,223
This is a 12VDC 1/2” throw 8 watt continuous duty push type solenoid. It is going to push up an object and keep it lifted permantently except for occasions where it turns off for a minute at a time which would be once a day or maybe not at all for months. I would assume that a 1amp supply is more than sufficient. It does get hot when on. Does anyone have experience with solenoids? My question is how to get the longest lifespan from both the device and the power supply. I can provide a separate power supply for this. The Prop turns on a logic level n mosfet that is rated far above 1 amp. Just looking for advice on whether to use some much higher rated PS for longer life. Also does continuous heat mean this thing fails at some point.
1124 x 1500 - 197K
Comments
https://www.mouser.com/ProductDetail/Texas-Instruments/DRV110PWR?qs=sGAEpiMZZMslbOXsVR1ov%2bWSD/S1HVPL
Or with built-in mosfet:
https://www.mouser.com/ProductDetail/Texas-Instruments/DRV103U?qs=sGAEpiMZZMtKB4wrjsn3lbJeWO/rMbZert/wGSc7LX0=
Or you could try to simulate that with P1, but maybe with this chip maybe don't need a mcu at all?
Better to make some sort of latching mechanism, and perhaps two solenoids, one to open and one to close.
Unless it is safety related. Energised to permit movement, de-energised to inhibit.
Yes.
The Force-distance curve for solenoids (which includes relays) is of course non-linear.
The holding current is much lower than the current needed to move when fully open.
A regulated current drive is better than voltage, as the Amp-Turns is what sets the force, and that removes the solenoid temperature as a variable.
Some designs sense the plunger position, and drive lowers when closed, and that is accidental release safer.
Others idle at PWM and then every 5-50 seconds drive at full power for a brief pull-in-time. That gives drop-out recovery that is not as immediate, but still 'somewhat there'
In that case, you probably need to brown-out test it, and the periodic full-pulse might be a good idea to add, easy enough if you have control of the SW
I went looking just for grins. Always on solenoids are a thing.
The document I linked recommended a power resistor and switch be used to limit holding current.
Seems like pwm would do the same thing.
Also mentioned a diode to limit back current.
Or if you need cheap, use a door lock actuator. Internal rack & pinion. Just drive it for a second in either direction and shut off. Stays where you put it, and manually overrideable. $2.84 https://www.ebay.com/itm/6EB3-Auto-Car-Vehicles-Central-Locking-System-2-Wire-Door-Lock-Actuator-Motor/253780948666
Better hurry. They had 3 and I bought one. Only two remain... and my mouse finger is itchy, those Chinese tariffs may be coming...
.. unless this is part of the spec... (see above)
"Must be fail safe. Energized to stop movement, de-energize to allow movement. So if no power for any reason it is free to move. Gravity drops the solenoid pin."
Servos fail there, tho some designs I've seen try to fudge things with super-caps and claims those have enough energy to unlock...
All gets complicated, and of course does not actually fail-safe at all, in all failure modes.
We had a situation where a machine supplying 15V was powering a 12V solenoid ... for the most part this worked fine, but the solenoid did get hot and fail prematurely. ...I did some empirical testing and found that our 12V solenoid would reliably work down to 4V (33% of the rated voltage). This got me thinking and we ended up changing the 12V solenoid to a 24V solenoid on all of our machines company wide. The 24V solenoid powered from a 15V supply was perfectly happy at 62.5% of the rated voltage of the solenoid.
All in all, my suggestion of using a spring to hold the object up was correct --- it is nuts to burn up electricity, and wear out your solenoid, to hold the object up for hours or even days at a time --- a spring will hold it up, and the spring won't wear out for many years.
Motors with integral brakes need to energise the brake's solenoid to allow the motor to rotate and will attempt to lock the rotor in the event of a power failure.
The company we use who manufactures these is ITC products, and they have a range of styles but not a huge amount of information on the net. There must be other similar manufacturers out there
Based on about 3 decades experience, in general they are very reliable (almost never need replacing) despite being in some challenging environments. The one clear exception is a "coal loading terminal" that juts out into the sea, they seem to go through a couple per decade.
Hope this helps.
https://www.mcmaster.com/70155k121
I think I disagree, I think if a holding duty cycle will produce lower heat that it will add to life.
If this was my problem to solve I would:
- measure current draw when full voltage is applied (Ifv)
- measure minimum current required to hold the solenoid in the up position (Ihm)
- use a power supply rated at 2 x Ifv
- add a constant current circuit that supplies a current half way between Ifv and Ihm
- add a parallel circuit and a 555 timer to provide full voltage to lift the solenoid
I would start with something like 1KHz. I seem to remember this being the carrier frequency of a Vickers proportional valve driver.
Will any one hear this ?
We used 19kHz, (just ultrasonic), with deliberately soft edges on the FET, via a largish 560R gate resistor, and a fast switching catch diode.
If there is a power failure you want your machine to revert to a safe state --- I mean, safe in the sense that the operator has minimal chance of getting injured --- the machine should stop moving when it is no longer under control.
For example, brakes would be held open by the solenoid when the machine is running, but the spring would clamp down the brake when power is off.
The OP would have to tell us more about what he is building to say which method is best.
I am going to set up a mosfet off the prop and test various frequencies and see how it behaves. I looked online at constant current methods and see the lm317 being used quite a bit. Very simple to make. The device is pulling .666 amps.
That current is pushing things for linear regulation.