# need help with reducing DC voltage

Discussion in 'General Electronics Discussion' started by GeoNOregon, Feb 27, 2014.

1. ### GeoNOregon

14
1
Jan 30, 2014
I'm a beginner in electronics, so please keep things simple.

I'm re-designing an air pump for a Sleep Number bed. It uses electro-magnetic solenoids to actuate valves to control the inflation and deflation of the two sides of the bed. It is a very poorly designed system and is plagued with leaking problems.

Instead of solenoids actuating valves in a leaky plastic box, I have designed a system of electro-magnetic air valves to control inflation and deflation. The OEM solenoids are custom designed and I can't get the specs. I fired up the pump and measured the voltage from the circuit board to the solenoids and got a 100v DC measurement.

Unfortunately, I can only get air valves in 120v AC, 12v DC or 24v DC. It seems to me, (based on my limited knowledge and experience), that I should be able to utilize the 100v DC outputs from the circuit board by reducing the voltage to 24v DC using resistors.

Is it as simple as computing the voltage differential between the 100v and 24v and then calculating the correct resistor to 'consume' the 76v difference? Or is there something else I need to do?

Thanks,

GeoD

2. ### (*steve*)¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥdModerator

25,491
2,833
Jan 21, 2010
No, it's not that easy unfortunately.

You may need a separate power supply for your valves and design some circuitry to trigger this based on the 100V signal.

3. ### duke37

5,364
771
Jan 9, 2011
The 120V AC valves will probably work on DC. You may need to drop the DC voltage to around 50V with a resistor equal to the valve resistance.

I have some 240V AC Norslo relays here.
They pull in and drop out at 130V AC.
They pull in at 60V DC and drop out at 7V DC.

Some experimentation would be needed.

4. ### kpatz

330
88
Feb 24, 2014
The simplest way may be to use the 120V AC valves and switch them using relays (like the ones duke37 mentioned) from the 100V DC controller outputs.

You'll need resistors that will drop the DC voltage from 100V to whatever the relay coil DC voltage rating is. The resistance needed will depend on the DC resistance of the relay coil and thus the current draw. With those two variables you can use Ohm's Law to calculate the resistance required.

You'll also need to use resistors with a sufficient power (watt) rating so they don't overheat. Something else that is simple to calculate once the voltage drop and current are known.

One other thing: try disconnecting the solenoid from the controller and then measure the voltage. It may be higher without a load. You did confirm it's a DC and not AC voltage, right?

5. ### GeoNOregon

14
1
Jan 30, 2014
screwed up now

I was going to address each suggestion and I had some further questions, but kpatz's comment about measuring the solenoid voltage without a load and if I confirmed it was AC or DC got me questioning myself.

I checked the solenoid voltage without a load and it ~104v, it had been a consistent 100v with a load.

On the question of being AC or DC, I thought my VOM wouldn't read AC when set to DC and wouldn't read DC when set to AC. To confirm that, I checked the line voltage to the pump assy. It comes from the wall, so it is ~120v AC.

On AC, I got 118.4v, on DC, it jumped around and wouldn't settle on a final reading and was '-' no matter which way I set the test leads of the VOM.

I checked the feed to the same solenoid was before and got the ~104v while set on DC, as I was switching the VOM to AC to check again, the black lead on the VOM, which was on the negative side of the solenoid supply leaned over and touched the positive lead of the deflate supply. A small crackle ensued... no smoke, just a slight crackle. It wasn't more than a glancing contact, not a steady one.

The pump no longer works, I can't get the remote to initialize; no voltage except at the line feed. So, I ****ed up the circuit board.

help...

6. ### kpatz

330
88
Feb 24, 2014
Ouch... sorry to hear that. Hopefully you just blew a fuse and replacing it will get you up and running again.

On the other hand, now you can build a new control board that supplies the voltage your valves need instead of working off the 100V supply. Or see if your bed is still under warranty. "It just stopped working, honest!"

7. ### GeoNOregon

14
1
Jan 30, 2014
No chance on the warranty; the bed was a gift from a friend's family, given to me for helping the family with his estate.

Re: fuse, if you are talking about a fuse that look's like a conventional fuse, there is nothing like that on the circuit board.

As far as building a new control board, that is so far beyond my capability/experience in electronics, that it's not feasible.

I'm not scared of trying anything, but I do know my limitations. I'd need much more hand-holding than I'd think anyone is willing to give. I've had a brain-injury and my capacity to learn new things is slower than it used to be.

What really fries me is I'm disabled from numerous injuries and haven't slept in the same bed as my wife in 12+ yrs. The \$4000+ Sleep Number bed I was given is way beyond anything we will ever be able to afford and offered a chance to sleep in the same bed, again. Oh well.

8. ### kpatz

330
88
Feb 24, 2014
Maybe look on eBay to see if anyone has a replacement control board for sale. Can you post a pic of the board? One of us might recognize a fuse that you don't.

9. ### GeoNOregon

14
1
Jan 30, 2014
Here's a pic of the board.

I shorted between P4 & P5 at the bottom of board. P3 & P4 are labeled 'Right' and P5 & P6 are labeled 'Deflate'.

I noticed the square, rust colored component as I was editing the photo. It is marked F1. I thought it might be for fuse and it IS in the line voltage circuit. So, I checked the continuity of it and it is open. Could that be the culprit that popped?

Last edited: Feb 28, 2014
10. ### davennModerator

13,813
1,945
Sep 5, 2009
you may be lucky to have got away with that as the only damage

Dave

11. ### GeoNOregon

14
1
Jan 30, 2014
It looks to my unknowledgeable eyes to say 500mA on the top of the fuse. For a permanent fix, I need to find a replacement with the same specs.

Could I get a away with substituting a different fuse of higher amperage just to see if the rest of the circuit & components function?

12. ### davennModerator

13,813
1,945
Sep 5, 2009

not unless you want to kill the rest of the circuit if there is yet another fault
a fuse of given rating is used for a reason .... protection

get another 500mA fuse or 2

cheers
Dave

Last edited: Mar 1, 2014
13. ### GeoNOregon

14
1
Jan 30, 2014
Ok, I got lucky. I picked up some 500mA fuses yesterday and replaced the popped one. Everything is working again.

Now, on to the previous relay discussion.
@ kpatz: re: confirmation of the source voltage from the circuit board being DC, as I said in a prior post,

"I thought my VOM wouldn't read AC when set to DC and wouldn't read DC when set to AC. To confirm that, I checked the line voltage to the pump assy. It comes from the wall, so it is ~120v AC.

On AC, I got 118.4v, on DC, it jumped around and wouldn't settle on a final reading and was '-' no matter which way I set the test leads of the VOM."

I then checked the solenoid voltage without a load on the DC setting and it was ~104v, it had been a consistent 100v with a load. I then set the VOM to AC and got a reading of 51.5v

Based upon the measurements results on the 120v AC, I think I can deduct the circuit board voltage to the valves is DC and not AC.

@ duke37: When I found out the available valves had voltages of 120v AC, 24v DC and 12v DC, I did some research on using AC coil-actuated devices with DC voltage thinking about using the 120v AC valve. I didn't find anyone recommending it or saying it could be done, and no information like you gave on the 240v AC Norslo relays.

@ kpatz, duke37 & (*steve*): What confuses me is if dropping the voltage to 50v to drive relays which would switch the line supply of 120v AC or the voltage of a separate power supply, why can't the voltage be dropped to 24v and just drive the valves directly?

Is it too much voltage drop, too much heat generation, can't get resistors to handle it or....?

14. ### duke37

5,364
771
Jan 9, 2011
I think the only way of finding out the power available is with an oscilloscope or studying the circuit.. It could be DC, AC, full wave rectified AC or half wave rectified AC.

Obviously, using a high voltage supply and using a resistor to drop to a low voltage involves a lot of lost energy and resulting heat.

If you have a 12V DC valve, get a 12V AC transformer and full wave rectify it. I do not think that any smoothing is required.

15. ### Gillesfizzog

18
0
Jan 14, 2014
Its possible. You need to pulse the dc current. Once it is pulsating dc, it can be transformed to 24volts. Than a run capaciter should be used to smooth the peaks of the pulsating dc. Some adapters used in car outlets do just that.

16. ### kpatz

330
88
Feb 24, 2014
EDIT: never mind, I didn't notice the 2nd page of posts before I replied.

Good to hear the fuse took care of things.

You could reduce voltage with a resistor, but heat dissipation could become an issue depending on the current draw of your valves. If they only draw a few mA of current, it won't be a problem, but if they're drawing 100+ mA you're going to need a big resistor. That's why I suggested a relay and just switch 120V AC to a 120V AC valve. Minimal fuss.

Do you know how much current your 12V and 24V valves draw on DC?

Last edited: Mar 2, 2014