PDA

View Full Version : Boost up current output



o_wan
03-04-2005, 02:20 AM
hi,

I realiseed that each pin can output 25mA and 75mA in total for basic stamp. By using a LED driver, I can get 8-16 (or more, it depends on which driver) individual 25mA outputs.

I've seen some websites which teaches how to boost up current by usign using transistors. Is there anyone experience in that?

I would like to have around 80mA, is that possible?

Cheers,
A

steve_b
03-04-2005, 02:40 AM
Here's a pic using a stamp pin to trigger a relay.

The stamp pin can't source the current required by the relay coil to trip...so a transistor is used.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·

Steve
http://ca.geocities.com/steve.brady@rogers.com/index.html
"Inside each and every one of us is our one, true authentic swing. Something we was born with. Something that's ours and ours alone. Something that can't be learned... something that's got to be remembered."

kb2hap
03-04-2005, 02:51 AM
you could use a uln2803 then each pin would give you 500ma you can look a a previous thread Darlington to read more

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
DTQ

o_wan
03-04-2005, 07:42 AM
hi,

what if I want the exact current output, is the uln2803 still suitable? Can I control the current output individually?

or I need to go for particular transistor?

I intend to hook it up with this biometal thingy, is there anyone experience on it with BS?

Cheers,
A

Orion
03-04-2005, 08:24 AM
Using my magic wand and Google I found the data sheet @ http://www.ozitronics.com/data/uln2803.pdf (second hit) from the data sheet FIRST 3 lines:

"Featuring continuous load current ratings to 500 mA for each of the drivers, the Series ULN28xxA/LW and ULQ28xxA/LW high voltage, high-current Darlington arrays are ideally suited for interfacing between low-level logic circuitry and multiple peripheral power loads. Typical power loads totaling over 260 W (350 mA x 8, 95 V)"

Read the datasheet

kb2hap
03-04-2005, 08:53 AM
Depend what you want to do the 2803 has a max value of 500ma per pin
so you can use anything within that range
if you want to do something variable you may want to use a digital potentiameter
or pulse width modulation

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
DTQ

KenM
03-04-2005, 11:08 AM
I get the impression from one of your posts that you want an "exact" current to go through your 'bio metal' thing, is that correct?

You want to force about 80 mA through the device, not 50 mA, not 100 mA, but a fixed amount......~ 80 mA, is this what you want to do?

If so, what you are looking for is a current source, and if this is the case is ~80 mA the only current you want, and the flow of current would be controlled by the stamp?

Ken


Post Edited (KenM) : 3/4/2005 4:10:50 AM GMT

o_wan
03-04-2005, 05:38 PM
yup, precise. I want to have individual current output at 150mA.

My question is

Is that possible to uln2803 to control current output individually?

If yes, any clues how?

Cheers,
A

KenM
03-04-2005, 05:52 PM
If you want the current to be controlled at 150 mA not much more, not much less, a current source is what you are looking for.

It is way too late for me to come up with something now...but for starters, you might want to google "constant current source from a transistor pair"

The uln2803's are not going to do what you want (or at least, I don't know how to do it with that device). I will post some possibilities Friday or Saturday, which is the best I can do for now.

Also, you might want to move this thread to the sandbox, as·this subject is not a basic stamp question.

For starters, take a look at this

Toshiba Constant Current LED Drivers (http://www.marktechopto.com/products/listing.cfm?drill_level=product&DeptID=2100&SeriesID=1286&Part_Number=TB62715FN)

Ken


Post Edited (KenM) : 3/4/2005 11:00:56 AM GMT

o_wan
03-04-2005, 09:08 PM
hi,

Thanks a lot for all your reply, at least I know it's do-able right now.

Is there anyone working with that Toshiba Driver before? Any suggestion?

Cheers,
A

Nate
03-04-2005, 09:59 PM
If you want a constant current to pass through the "BioMetal" device,·and you have a known constant voltage·available to the transistor, you just need to put the proper resistance in series with the device.· By ohms law your current would be defined.

This seems so obvious, I think I·must be·misunderstanding your question· :).

Nate

Edit- Is the resistance of the BioMetal not constant or linear?

Post Edited (Nate) : 3/4/2005 3:06:10 PM GMT

KenM
03-04-2005, 10:18 PM
Edit- Is the resistance of the BioMetal not constant or linear?
I briefly looked at the device, it appears the resistance is non-linear, hence the need for a constant current source.

k

steve_b
03-04-2005, 10:35 PM
What happens when temperature changes and affects your transistor and resistor and such....voltage levels change and then the current going through the resistor would change also....

Anyhow...just threw together a circuit out of a book.· Might help!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·

Steve
http://ca.geocities.com/steve.brady@rogers.com/index.html
"Inside each and every one of us is our one, true authentic swing. Something we was born with. Something that's ours and ours alone. Something that can't be learned... something that's got to be remembered."

o_wan
03-04-2005, 11:27 PM
Apart from the transistor problem, I'll need to drive not only 1 biometal but at least 8 of them, that's why I think I'll need a driver , something look like LED driver, a constant current mumltiplexer.

allanlane5
03-04-2005, 11:42 PM
A 741 op-amp IS a 'driver'. I thought the constant-current design above was quite elegant.

And an LED driver is NOT a 'constant current mux', typically it's an open-collector transistor which gives you a low-resistance path to ground when turned-on -- very similar to the 'darlington' approach you opened this discussion with.

The nice thing about a constant current source is if you hook things in series, the current goes through all of them. That's true until the resulting voltage gets too high, of course.

"constant-current" through a changing resistance is not a trivial problem.

Jon Williams
03-04-2005, 11:46 PM
Does you biometal thingy change the electrical characteristics of the circuit when it's activated? If so, enough to require a complex circuit? If not, perhaps a clean regulated supply and the proper resistor in series with it (Ohm's Law is our friend) is enough to enlist the ULN2803. Call me simple, but simple usually works best.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jon Williams
Applications Engineer, Parallax
Dallas, TX· USA

SteveW
03-04-2005, 11:49 PM
Some recent LED drivers do deliver constant current - it's the only way run cheap, high brightness LEDs with a consistent brightness for backlighting.
Maxim's MAX6977 , f'rinstance. Only 50mA so no use for this job, but... (and others in the family can tolerate higher voltages)
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/4671


Steve

Bruce Bates
03-05-2005, 12:15 AM
Folks -

Just to add a bit more information to this ongoing and somewhat confusing thread, this "bi-metal thingy" or "bio-wire" is actually a form of SMA (shaped metal alloy) similar to Nitinol, "muscle wires" or call it what you will. The "suggested current" for the particular gauge wire that is being used here is 150 ma. The resistance of the wire is 375 ohms/meter.

Hopefully that will put a better perspective on things.

Just as in passing, during this research I came to find out that The Robot Store (aka Mondo-Tronics) has been purchased by Jameco, as noted on their web site. The Robot Store is a source of various forms of SMA products. The can be found here:
http://www.robotstore.com/

Regards,

Bruce Bates

Paul Baker
03-05-2005, 01:06 AM
http://www.edn.com/article/CA446990.html

shows a constant current, constant voltage driver circuit using an LT1618, who's spec sheet is here: http://rocky.digikey.com/WebLib/Linear%20Tech%20Web%20Data/LT1618.pdf

it can be controlled simply by the stamp using the /SHDN pin. One note: its only availible in the MSOP package which is surface mount .5mm pin spacing.


Post Edited (Paul Baker) : 3/4/2005 6:13:41 PM GMT

o_wan
03-05-2005, 02:16 AM
thanks a lot for the reply, since I was working on max7218 LED driver, I think I'll go for the ULN2803, it's cheap and should work for my purpose.

cheers,
A

allanlane5
03-05-2005, 02:29 AM
Having looked now at the website, and the (translated from Japanese) "documentation", I think the Darlington will do what you want -- but you WILL have to put a resistor in the circuit to limit the current through your 'spring'.

The 'Bio' in there may have thrown some of us off -- basically this is a piece of wire, wound into a very small coil, which has the property that you can stretch it at room temperature, then it will return to its original shape when heated by current going through it. It will start returning to shape at 20 mA or so -- 150 mA is the MAXIMUM you can use, which will probably shorten the life of the device. So size your resistor (and include the resistance of the spring) so you get 50 mA, and see what happens.

This is a 'robot muscle', people, that's all.

Update:· Oops, depending on the diameter of the wire, 150 mA or 220 mA are "Typical" currents for the big stuff.· My apologies.


Post Edited (allanlane5) : 3/4/2005 7:35:56 PM GMT