I need to switch some GE #44 lamps from a TTL output. #44 lamps are 6.3 volts at 250 ma, so I'll need a transistor to do the real work. Unfortunately, it's just a hair too much power for one of the hundreds of surplus 2n2222's I keep around to do almost everything else here. Can someone suggest a small, cheap available transistor that would be appropriate for this use?
-- Lee K. Gleason N5ZMR Control-G Consultants snipped-for-privacy@comcast.net
Use a FET? The NDS351 comes to mind but only if it's driven from 5V TTL logic. Around $0.10 in large qties. If it's 3.3V logic you'll have to keep looking for one that has guaranteed Rdson data down there and is low enough in cost.
formatting link
--
Happy New Year, Joerg
http://www.analogconsultants.com/
For switching DC voltage -- If the transistor is driven straight from a non-open collector IC -- I suggest using a high gain transistor such as MPSA13, MPSA14 or 2N6427 for switching one lamp. If switching two lamps -- MPSU45, CENU45, 2N6548 or
2N6549. None of these four are real popular, though. The CENU45's are still made - other three are tough to find.
If the transistor is driven from an open collector IC with a stiff pullup resistor on the output then you can use something like a 2N6037 for switching one or more lamps.
If the transistor is pre-driven by a pre-amplfication transistor (i.e.
2N2222) -- then you can also use something like the 2N6037.
As an alternative, you can also use logic level driven MOSFET's as well -- something like the IRL510 would drive lots of lamps.
If you're switching AC voltage than use something like the Bally pinball machines used -- 2N5064 for one lamp or MCR106 for more than one lamp.
That would need to be a pretty high saturation voltage in the transistor to get 1.5W at 250mA.
The MPSA13 is rated at 625mW. With a saturation voltage of 1.5V and 400mA (typical GE44 is actually about 400mA) -- you're dissipating 0.6W... maxing out the MPSA13. Pinball people have been using MPSA13's and GE44 bulbs for years. The biggest problem happens when the bulb burns out. Unpredictable things happen within the bulb when they burn out -- sometimes the filliment will break, move and 'reweld' itself to a different place in the filliment -- giving a brighter bulb...and short lifespan on the transistor.
You can also go with a type 47 bulb. Same voltage but about 250mA. Less current so less power dissipated by transistor. Down side is the bulb is considerably dimmer.
The little detail is that "beta=10" thingamjig...that means if you drive the base with 25mA, the Vce at 250mA will be less than 0.2V. This makes the Pd = 0.25A x 0.2V= 0.125W. Now look at figure 3, Vbe,sat, at 250mA is about 0.9V. Use this to compute the voltage bucking your TTL drive. Then the base dissipation is 0.025A x 0.9V=0.0225W. The grand total is
0.0225+0.125=0.1475W. Now go up to the top and look at R theta,j-a, of
200oC/W for the TO-92 plastic, that gets you 200oC/W x 0.1475W=29.5oC junction rise above ambient. Now look at Tstg under max ratings, it says
150oC, this means you will be fine at ambients less than 150-30=120oC. recall this is 120x1.8+32=250oF. Which TTL part are you using to drive it? The 25mA base drive would suggest something called a *buffer*.
The initial current surge (maybe 10x of normal) of the light bulb will likely take out the transistor unless some sort of "soft-start" is implemented.
...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC\'s and Discrete Systems | manus |
| Phoenix, Arizona Voice:(480)460-2350 | |
| E-mail Address at Website Fax:(480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |
America: Land of the Free, Because of the Brave
I'm a sucker for the Zetex low saturation TO-92 transistors, but they are not very cheap and made by no one else. But drool over this data sheet:
formatting link
But if I was going for dirt common and rugged, I would probably go with a 3 amp rated power tab transistor mounted standing straight up on the board.
Gain is more important than current rating, but look for something with its peak gain in the 400 mA area or above, for low base drive and saturation voltage. That puts you up in the 3 to 6 amp rated sort of thing.
The TIP 31 has higher gain at 400 mA than the larger TIP41 has, so it is probably better, given the limited base drive.
formatting link
formatting link
This one is pretty good at 400 mA, also.
formatting link
But for my own stuff (not volume production design) I love the Zetex. Having 3 or 4 times the current gain really simplifies base drive and thermal problems.
Bah. 2N2222 is the same thing as 2N4401, and I've got a 2N4401 in a one-cell-powered-LED-inverter (apparently called a "Joule thief") running as a blocking oscillator around 5MHz, with peak emitter current around
800mA. Saturation is under 0.3V. Okay, probably half of that is base current, but it's still more than low enough.
Use a PNP (2907/4403) to take advantage of the TTL's current sinking capacity.
Tim
-- Deep Fryer: A very philosophical monk. Website @
So every HOT ever made isn't at all reasonable (hFE typ 2.5 to 5)?
:^)
Storage time sure goes to hell, though. Even fast HOTs are a couple microseconds. ZTX651 is something like 800ns at beta = 10 (a good 10-20 times less than hFE at the same Ic).
At any rate, 2N2222 has a fat junction, and is specified for up to 500mA, so there.
Tim
-- Deep Fryer: A very philosophical monk. Website @
John Popelish wrote in news:1dqdnR4N2YlibOfanZ2dnUVZ snipped-for-privacy@comcast.com:
Cool, someone mentions Zetex. They do seem to have a uniquely strong spec for such small transistors. Even the tiny e-line thingers make some TO92 types look clumsy. Maybe the small size radiates heat better, less bulk to seep through the way it must out of a TO92. For switching though, I'd still prefer MOSFETs. Many will have VERY low RDS with a 5V gate input.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.