Solenoid Matrix

Hello! I've been working on a project that will involve 42 solenoids in a matrix (3x14). The solenoids will be independent of each other, but the maximum number that will be on at any moment is 28. They should remain up until the next pattern is rolled on (after approximately 10 seconds). The solenoids are 3V with a current of 0.17A. I was wondering if an Arduino Mega will fit this task. Also, will it be able to supply the power needed or do I need an external source?

The main task of this device is to be a sort of Braille display.

Sure, Mega can supply the control signals, but not the current needed for that many solenoids.
For that you need a driver board.
This one I offer can do 32 channels of current sinking and can be daisychained for more.
http://www.crossroadsfencing.com/BobuinoRev17/


And it only needs an Uno to send out the 4 bytes per card.

I also have a P-channel MOSFET board if you need to Source current (provide a High output) vs Sink current (provide a Low output).

You Americans :o

32P-MOSFET_board_populated.jpg

Nice boards, but OP only wants to drive small solenoids.
Something like a TPIC6C595 can drive eight small solenoids directly.
No other parts needed, not even diodes across the solenoids.
Well, only some power supply decoupling caps.
6 of those chips can be daisey-chained.
Only three control lines, so any small Arduino can be used.
Leo..

TPIC6C595 is good for 100mA, not 170mA.

TPIC6A595 can do 350mA, that's what would be needed.
I don't have a board with those chips yet.

42 solenoids, need 6 chips. $3.87 each.
http://www.digikey.com/product-detail/en/texas-instruments/TPIC6A595NE/296-9007-5-ND/378481

LarryD, nice job doctoring up my picture! Very clever :slight_smile:

Ok, so I'm picturing a card like this. Diodes needed as the TPIC6A595 does not include them.

CrossRoads:
Diodes needed as the TPIC6A595 does not include them.

I think it has been mentioned in another recent thread that all these chips have active voltage clamping circuits.

From the datasheet, page 1: The device contains a built-in voltage clamp on the outputs for inductive transient protection.

No diodes needed.
Leo..

Wawa:
I think it has been mentioned in another recent post that all these chips have active voltage clamping circuits.

From the datasheet, page 1: The device contains a built-in voltage clamp on the outputs for inductive transient protection.

No diodes needed.
Leo..

I don't see any.
Refer to a ULN2803 pin 10 to see how that would look if they did supply kickback diodes.
You need to add them externally.
.

Voltage clamp != kickback

There is a 33volt or 50volt zener diode (depends on which IC is used) between drain and gate of each internal output mosfet.
If the drain rises above that threshold (from the solenoid kickback), the mosfet gets gate drive and starts conducting. Stopping the drain from rising any further.
An active system like this has faster solenoid/relay release times than a commonly used kickback diode.
Leo..

And this for a board layout.

Wow, you whip them up fast.
Needs some polishing though (and removal of the diodes :grinning: ).
I usually spend a day designing boards like this (don't like eagle's auto-router).
And another day polishing it.
Leo..

I am making the board to allow for diodes, a user does not have to install them.
I don't see anything on page 4 on the outputs that will shunt high voltage spikes back to the motor/solenoid supply.

Still needs some naming added to the 6-pin headers, and the diode names cleaned up.

Needs some polishing though (and removal of the diodes :grinning: ).

If we agreed with you we would be 100% wrong!

.

CrossRoads:
I don't see anything on page 4 on the outputs that will shunt high voltage spikes back to the motor/solenoid supply.

The picture "TYPICAL OF ALL DRAIN OUTPUTS", page 4, shows a zener from drain to gate.
When that zener/diode starts conducting due to a spike on the drain, the fet starts conducting.
Kickback energy is directed to source/relay ground.

The TPIC6C595 datasheet, page 3, puts a 33volt value on the zener.
This is 50volt for the TIPC6A595.

Datasheet states clearly on page 1: Power driver applications include relays, solenoids, and other medium-current or high-voltage loads.
Leo..

So a 50v spike when the solonoid is turned off. Hence . . . noise.
That's why you place a kickback diode across the inductance, you get 0 volt spike, oh my.

.

Please take a look at this solenoid - does it seem like it needs current control in both directions? Don’t they normally just extend (or retract) when energized and then spring action returns them to the unenergized state?

http://www.amazon.co.uk/gp/product/B01481020Y?psc=1&redirect=true&ref_=ox_sc_sfl_title_1&smid=A50DZI580G3JX

  • When the orange red line is connected to the positive, the white line is connected to the negative, offset magnetic.When the white line is connected to the positive, the orange red line is connected to the negative,enhance magnetic.
  • When the voltage is DC 3V, the current is 0.17A, it can not offset the magnetic completely.When the voltage is DC4.5V, the current is 0.28A, it can offset the magnetic,but it also overheats easily.

May be wrong, but this version appears to rely on the external connection to have the return spring or maybe just gravity.
.

Seems they have a magnetic plunger, so it can move in both directions.
Leo..