Operating 120 VDC solenoid from higher voltages

MisterBill2

Joined Jan 23, 2018
18,176
I may be barking up the wrong tree but if the SMPS operates from a 3 phase mains supply it must have a neutral as one of the inputs.

eg. as per this ST app note https://www.st.com/resource/en/appl...ications-with-viper12a-stmicroelectronics.pdf
No, there is no intrinsic need for a neutral connection from a 480 volt 3-phase source. and if the arrangement is a DELTA connection there may not even be a neutral available. So the best choice will be a transformer to step down the 480 volts to an isolated lower voltage and use that isolated source for the DC supply for what I am guessing is an external trip coil on a circuit breaker.

BUT, as often happens, we have not been given enough information to do better than make guesses. The result is thus a collection of guesses based on incomplete information. WE have only a voltage and resistance for the solenoid, giving a current of about 120/290 amps, and that it will be triggered by "a micro", which means that the isolation must be good and that the trigger will probably be a 5 volt signal that is referenced to some sort of low voltage arrangement.
 

Thread Starter

Blue_Electronx

Joined Jun 10, 2019
112
I don't need an exact value, just the minimum expected time between consecutive operations of the solenoid.

But after further looking at possible circuits, it will still require quite high power resistors, even with a low duty-cycle, so you may have to consider adding a transformer to drop the AC voltage down to about 90V to generate a rectified 120Vdc.
I don't have space for a transformer unfortunately. What do you think about some sort of current control with a shunt resistor?
 

MisterBill2

Joined Jan 23, 2018
18,176
I don't have space for a transformer unfortunately. What do you think about some sort of current control with a shunt resistor?
You could use a fairly high value resistor and diode to charge a capacitor and then use an SCR to dump the charge through the solenoid. The trigger for the SCR can be a trigger transformer which is small.
But if tghere is room for some micro board and that micro board needs a 5 volt supply then you have room for that. Also, NOTHING associated with 480 volts is small. AND, if you have 120 volts for the supply for the micro then it can feed the DC supply for the solenoid as well. So we still are looking for the rest of the information.
 

rherber1

Joined Jan 6, 2008
27
Why not use a DC-DC converter?
I was about to suggest the same thing... but then I thought better of it.

The DC-DC converter would probably have to run off the 12V output which is used for the logic (the OP gives no details regarding the SMPS he is using) so the current rating of this supply would probably be inadequate. Since the solenoid requires 50W while energised (assuming it is not an AC solenoid), the converter would need to be rated at least for this power output (and most probably 100W) which means the converter would need to draw in excess of 4A from the 12V supply just to get the 120V required to energise the solenoid.

Also, the converter would take up considerable space - which the OP presumably doesn't have to spare.

I fear the OP is going to be disappointed yet again - but then he hasn't been particularly forthcoming or helpful in his requests or answers.
 

MisterBill2

Joined Jan 23, 2018
18,176
I was about to suggest the same thing... but then I thought better of it.

The DC-DC converter would probably have to run off the 12V output which is used for the logic (the OP gives no details regarding the SMPS he is using) so the current rating of this supply would probably be inadequate. Since the solenoid requires 50W while energised (assuming it is not an AC solenoid), the converter would need to be rated at least for this power output (and most probably 100W) which means the converter would need to draw in excess of 4A from the 12V supply just to get the 120V required to energise the solenoid.

Also, the converter would take up considerable space - which the OP presumably doesn't have to spare.

I fear the OP is going to be disappointed yet again - but then he hasn't been particularly forthcoming or helpful in his requests or answers.
A DC to DC converter is the last choice, and probably the worst one. Given that the on time for the solenoid is about 25 milliseconds, it would work to use a series resistor of not that great a wattage. Since the purpose is claimed to be tripping a shutoff when there is a fault caused overload on the 480 volt circuit it should not be operating very often. And the claim that there is no room for a transformer seems totally unreasonable, given that there seems to be room for an arduino and the stuff that it needs to interface with anything.
In addition there is a response time issue, since fault protection needs to be fast, of course the arduino running some program is not likely to be nearly fast enough. I wonder iif the TS has considered that.
 

crutschow

Joined Mar 14, 2008
34,283
it would work to use a series resistor of not that great a wattage.
For rectified DC from the 480V, the series resistor instantaneous power would be over 200W for the pulse period so, even though the average power may be low, you would still need a pretty good sized resistor to handle that transient value.
 

Thread Starter

Blue_Electronx

Joined Jun 10, 2019
112
In addition there is a response time issue, since fault protection needs to be fast, of course the arduino running some program is not likely to be nearly fast enough. I wonder iif the TS has considered that.
Yes, fault protection needs to be fast. I'm not using Arduino, I'm using STM32. Why wouldn't it be fast enough? A lot of breakers, and fault protections devices nowadays are based on microcontrollers.
 
Last edited:

Thread Starter

Blue_Electronx

Joined Jun 10, 2019
112
T
I was about to suggest the same thing... but then I thought better of it.

The DC-DC converter would probably have to run off the 12V output which is used for the logic (the OP gives no details regarding the SMPS he is using) so the current rating of this supply would probably be inadequate. Since the solenoid requires 50W while energised (assuming it is not an AC solenoid), the converter would need to be rated at least for this power output (and most probably 100W) which means the converter would need to draw in excess of 4A from the 12V supply just to get the 120V required to energise the solenoid.

Also, the converter would take up considerable space - which the OP presumably doesn't have to spare.

I fear the OP is going to be disappointed yet again - but then he hasn't been particularly forthcoming or helpful in his requests or answers.
The SMPS is intended to provide low current for the control circuit. In some cases even a capacitive dropper is used to get low current from high voltage. So a 1W SMPS would be enough. The output is 12V/100mA. However, this could be changed. I just don't think I should get away with a high wattage/expensive power supply. In old designs I have seen this kind of solenoid being operated from the high rectified DC voltage using some mosfet/resistor combination to mitigate voltage stress. And as said in previous comments, the coil can be momentarily overdriven.
 

MisterBill2

Joined Jan 23, 2018
18,176
For rectified DC from the 480V, the series resistor instantaneous power would be over 200W for the pulse period so, even though the average power may be low, you would still need a pretty good sized resistor to handle that transient value.
The cheating trick is to have a quite high value resistor in series with the diode to charge the capacitor. When the circuit is activated to deliver the current only the first bit is the higher voltage, and that is when a solenoid needs the greatest current. The voltage quickly drops as the capacitor discharges and so the total watt seconds dissipated in the resistor is not so very great. Consider if we use that previously posted 200 watts, and that stated 25 milliseconds time, then 200W x 0.025sec =5 watt-seconds. Not that much heat. And the resistor used for charging could be half the value of the capacitor's leakage resistance, which is probably fairly high, since this application takes more than a cheap capacitor to provide the needed reliability.
 

MisterBill2

Joined Jan 23, 2018
18,176
Yes, fault protection needs to be fast. I'm not using Arduino, I'm using STM32. Why wouldn't it be fast enough? A lot of breakers, and fault protections devices nowadays are based on microcontrollers.
OK, that one is probably much faster. I just presumed an arduino because so many folks want to use them. So my guess was off. Sorry about that error.
 

crutschow

Joined Mar 14, 2008
34,283
And the resistor used for charging could be half the value of the capacitor's leakage resistance, which is probably fairly high, since this application takes more than a cheap capacitor to provide the needed reliability.
The tradeoff if charging time versus resistor dissipation, since the capacitor would likely need to be at least 50uF to operate the solenoid for 25ms.

And 50uF capacitance @ 800V capacitors is not small.
 

MisterBill2

Joined Jan 23, 2018
18,176
The tradeoff if charging time versus resistor dissipation, since the capacitor would likely need to be at least 50uF to operate the solenoid for 25ms.

And 50uF capacitance @ 800V capacitors is not small.
use two 100MFD/500 volt motor starting capacitors in series. No, not really small. OR, use a larger resistor and a zener shunt diode to limit the capacitor charge voltage to a lower value. The trade-off is a longer recharge time. Charge to only 350 volts and then use only one 400 volt starting capacitor.
 

crutschow

Joined Mar 14, 2008
34,283
use two 100MFD/500 volt motor starting capacitors in series. No, not really small. OR, use a larger resistor and a zener shunt diode to limit the capacitor charge voltage to a lower value. The trade-off is a longer recharge time. Charge to only 350 volts and then use only one 400 volt starting capacitor.
That could work if he has room for the capacitor and can tolerate a fairly long (perhaps a minute or so) charge time.

And you could likely use a smaller capacitor.
A 50uF cap charged to 350V would drop to about 60V at 25ms when connected to the solenoid, which probably is enough to keep it pulled in.
 

MaxHeadRoom

Joined Jul 18, 2013
28,619
A 100ufd motor starting capacitor is a bipolar electrolytic.
They can tolerate a long application time on DC, just a few seconds on AC.
The current required to retain a DC coil is a lot less than the pull-in current.
Max.
 

MisterBill2

Joined Jan 23, 2018
18,176
The application description sounded like a trip mechanism for a circuit breaker. A lot of non-residential breakers do have an external trip connection. But since that was not specifically called out, it is only my best guess. A 25 millisecond on time is not long by most standards. Shorter than most motor spin-up times for typical FHP induction motors.
 

Danko

Joined Nov 22, 2017
1,829
I have a 120VDC solenoid that I want to use in 480VAC application, i.e., rectified ~680VDC. The signal to activate the solenoid will be given by a micro. The solenoid won't be continuously activated, it needs just a pulse of about 25 ms. The resistance of the coil measured with a meter is 290 ohms. Is there a way, maybe using a MOSFET, to operate the coil at this higher voltage? Maybe the mosfet with the help of some resistor reduces the voltage stress on the coil?
For example:
With solenoid MAG-SP D45/207 110 D _ 32W,
110V DC, Rcoil=378 Ω, L=4600 mH, P=32 W,
pulse 240 ms:
1586126175447.png
 

Reloadron

Joined Jan 15, 2015
7,501
When all is said and done if I could not get a coil swap I would just get a small 480VAC - 120 VAC control transformer and put a bridge on it. Not like you need much power. A small 75 VA control transformer, a fuse and a full wave bridge. The last thing I ever wanted to see in a control panel was 480V.

Ron
 
Top