Smoking ULNs....

[JonnyMac]

Well, the season is officially here because I know of at least two stories having to do with smoking ULN chips.  The reason this happens is that too much current is drawn through the ULN, it overheats, blammo -- dead bug.  Unfortunately, however, the ULN can occasionally fail in a manner that does damage to the Prop-1 (or Prop-2); this is usually the result of a HUGE over-current on the ULN output.  I realize the that ULN specifications can be a little mysterious so I've tried to simplify things -- and err on the conservative side -- and put additional information vis-a-vis the ULN into the our prop controller documentation. 

Okay, first things first: the maximum current you should attempt to pull from the ULN is 1200 mA.  That's all pins in use added up; not per pin!  Per pin I suggest you limit the output to about 225 mA, but don't let the accumulative total of all pins exceed 1.2 amps.

The problem, of late, seems to be with unknown solenoids.  Perhaps we find them on eBay or other reseller of used products.  You will probably not find a current rating on the solenoid; what you should find, however, is a wattage rating.  For example, I was looking at a valve site and found one listed at 2.5 watts.  What does this mean as far as current goes?

We'll keep things simple and use the formula:

    P (watts) = I (current) x E (voltage)

Since we want to know the current we can rearrange the formula to:

    I (current) = P (watts) / E (voltage)

So if we use this valve in a 24 volt application the current draw will be 2.5 / 24 = 0.104, or 104 milliamps.  And at 12 volts?  208 milliamps. So if we want to control six of these valves it is better to run them at 24 volts to minimize the current through the ULN.

One last item: Solenoids and other inductors have what's called "inrush current" -- when the plunger is moving from its extended position to being centered in the coil the current is momentarily higher (in some cases considerably higher).  So if you are going to run a bunch of solenoids simultaneously you may want to consider sequential start-up instead of activating them all at once; even a short delay will help and will probably not have an adverse effect on your prop. 

Instead of this:

  DIRS = %00111111
  PINS = %00111111

You might want to do this:

  FOR thePin = 0 TO 5
    HIGH thePin
    PAUSE 5
  NEXT

In the end, if you find yourself with a solenoid that you're unsure of, give us a call; we'd rather spend a few minutes talking about it before you make the connection than after damage has been done....

[clinefx1]

I know "inrush current" wakes you up when you take it in the hand!


-chris

[JonnyMac]

And if your fist clenches around the live wire you may gain a whole new appreciation of "holding current"....

[clinefx1]

"holding current"  That?s the gift that keeps on giving. :o


chris

[Bean (Hitt Consulting)]

If you have a meter that can measure current (almost all can). Just connect the meter so that is is measureing current (sometimes you have to plug the red wire into a different hole), then with the ULN off, connect the meter from the device pin to ground. This will activate the device and show how much current the device is using.

If you have a 12V solenoid DO NOT RUN IT FROM 24V, you will smoke the ULN and the solenoid. You must use a 24V solenoid to get half the current as in Jon's example.

If you are controlling DC light bulbs, then you CAN put two 12V bulbs in SERIES connected to 24V and pull the same current as a single bulb connected to 12V. (Watch because light bulbs have inrush current too).

If you have something that needs alot of current use a IRF510 MOSFET. With a heatheat this can handle several amps.

Bean.

[livinlowe]

Bean-
"Heatheat"? ;D

[Bean (Hitt Consulting)]

Yeah, you know a heatheat. Very close cousin to the heatsink.  :D

Bean

[livinlowe]

Ohhh! Right, right,right......saaay no more!!!!!!   ;D

[PerfessorEvil]

Hi Jon,
Sorry to open an old thread, but I'm guessing that this is what I just did.
I have two of the 24vdc 4.5 watt solenoids (from those surplus box pushers everyone was getting for a while), and I set up the code to either fire one, the other, or both. 
They fired separately with no problems for a few cycles, and then started to fire at the same time.  Almost immediately, they reversed, and then nothing.

I took apart my control box, and the ULN was hot to the touch, and the red leds on the board were blinking instead of steady.  When the power switch is in position one, the red LEDs  are steady.

I'm guessing I burned out the ULN and the blinking LEDs indicate a short?

I'm using a 24vdc 1.67a center positive power adapter.

Is the ULN something I can get from Radio Shack, or will I have to order one?

Thanks,
Rik

[JonnyMac]

You know, 4.5 watts at 24 VDC is not a problem, even with two on at once.  By my trusty HP calculator 4.5 watts works out to about 190 mA; this is no problem for the ULN.  I'm wondering if you have shorted coils in the solenoids.  Can you measure the coil resistance with a meter?  Are you sure you're running the coils at 24 volts?  If you run them at 12 you'll demand 375 mA per channel and now two at a time will let out the blue smoke.

You won't find the ULN at RadioShack.  You can order them from us or from electronics companies like Jameco, Mouser, or Digi-Key.

[PerfessorEvil]

Ah heck... if a short can cause this, then I may have caused it myself out of sheer laziness.  I was just doing a quick test, so I just wrapped the wires around each other... no electrical tape, no wire nuts.  It's very possible that the motion from the pneumatic caused them to touch.

Hopefully that was a $2 plus shipping lesson, and not a $35 lesson.   :(

[JonnyMac]

We put the ULN in socket for a reason; I sincerely doubt that you've done any harm to the Prop-1.