PDA

View Full Version : 10 amp BS2 controlled variable power supply (0 to 12V)



Boombox
09-18-2006, 08:28 AM
I need some help coming up with a 0 to 12 volt 10 amp variable power supply circuit that I can control with a BS2. Any ideas??


Thanks!
Boombox

Desy2820
09-18-2006, 10:01 AM
I'm not an electronics expert, but I did find this on the National Semiconductor's website.· The basic idea is that you would use three LM338 adjustable 5 Amp regulators.·

http://www.national.com/pf/LM/LM338.html

Look for the link to read the datasheet on-line, then scroll down to page 10, under "Typical Applications", the one I saw was rated to 15 Amps.

Instead of R8, a 2K ohm pot, you'd subsistute a digitally controlled pot.· This device recieves a digital input and adjusts it's resistance.·

Then connect the digital pot to the BS2.

There are some catches.· First, the lowest voltage out would be 1.2 volts, not zero as you asked for and a minimum load of 100 milliamps.· (Although a relay controlled by the BS2 could disconnect the load.)· Second, with these regulators, you need about 1.5 volts input above your·output, at 12 volts out, you'd need at least 13.5 volts in...at about 11 amps.· Finally, your going to need a large heatsink for the regulators and will probably need to use a fan too.

If this idea doesn't work for you, then just search for "10A adjustable power supplies", "schematics" and combinations of these words.· You should get a huge number of hits.· The digital pot idea subsitution idea should work in most·designs that use a variable resistor.

I hope this helped!




Post Edited (Desy2820) : 9/18/2006 2:08:13 AM GMT

Loopy Byteloose
09-20-2006, 12:17 AM
If you have a 'bi-polar' supply [+10v and -3v] you can get all the way down to 0.

If you want to control the power supply's voltage with a BasicStamp, a digitally controlled potentiometer would do the job most smoothly.

Take a look at the PDFs from the manufactures for the LM317 and LM350.

Alternatively, you can have digitally enabled selections by using a 2n2222 to turn on and off different set-ups of the regulator's control pin.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"If you want more fiber, eat the package.· Not enough?· Eat the manual."········


···················· Tropical regards,····· G. Herzog [·黃鶴 ]·in Taiwan

Boombox
10-08-2006, 11:57 AM
Ok I built the LM338 circuit and it works great with a 10k pot. But when I attach it to a 10k digital pot (AD5220) it stays at about 6 volts no matter what. Any ideas?



Thanks!

Boombox

Mike Green
10-08-2006, 12:09 PM
There's either something wrong with the way you've wired things up or there's something wrong about the way you're controlling the digital pot. See if you can post a schematic diagram of what you've actually constructed (not what you think you've built) and we'll have a look at it.

Boombox
10-08-2006, 01:47 PM
The Power supply works great when I put a 10k pot in place of the AD5220. As far as the wiring of the AD5220 please see the attachment.

Mike Green
10-08-2006, 10:29 PM
There's no obvious reason why this should not work. The initial resistance value should be about 5K which is higher than you might want, but shouldn't cause the behavior you're seeing. I suggest you hook up the AD5220 separate from your regulator and use an ohmmeter to measure the resistance between the wiper and ground as you adjust it from the Stamp to make sure it works ok. I assume you have a proper common ground between the Stamp and the rest of the circuit.

Boombox
10-09-2006, 02:14 AM
Yeah, I thought of that to so I set up just one lm338 and an ad5220 and I get the same results. I've tried four different digital pots with the same results each time. Is it possible that it dosen't work because the digital pot can't sink as much current as a normal one? If so how do I get around that problem?

Chris Savage
10-09-2006, 04:03 AM
Boombox,
·
·· Doubtful…It should work the same for small signals…What voltage are you putting into the AD5220?· I would have to double-check, but for some reason I believe that will only work with a 0-5V signal on the internal resistor terminals.· You can verify from the datasheet.· I hope this helps.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support

Mike Green
10-09-2006, 04:19 AM
That may be the answer. I think the LM117 and probably the LM338 are designed to keep the output voltage about 1.2V above the adjustment terminal. If you're seeing about 6V at the output of the regulator, the AD5220 may be clamping the voltage at the wiper at about 5V. One solution could be to use a different digital pot. Another might be to use an op amp to buffer the adjustment terminal voltage so that a different voltage divider could be used, one that would make the adjustment voltage always less than 5V.

Boombox
10-09-2006, 07:45 AM
Sounds like the op amp is the ticket. which one would you recommend? 741?

PJ Allen
10-09-2006, 08:51 AM
Hate to ask, but have you actually tested the "digital pot" by itself --·and verified a·variable resistance?

N.B.·-- If the digital pot snuffs it, then will it go to max resistance or short?· If it goes short then the output voltage of the V.R.·will go to 0V (no problem), though if it goes to some high resistance then the output voltage will go to MAX.

Boombox
10-09-2006, 09:56 AM
Yup, I did·a full sweep and the digi pot looks good. What else can I try?

Chris Savage
10-09-2006, 10:59 AM
Boombox,
·
·· I still think it would help everyone to help you if you posted a schematic or a link to one so we could see what you’re building or trying to build.· Is the first link·that was·posted where you're getting your idea from?


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support

Boombox
10-09-2006, 11:09 AM
I'm sorry, yes it is. I'm using the LM338 15 amp schematic located on page 10 o f the link above. But I'm using the AD5220 instead of the 2k variable resistor (R8).



Thanks in advance!!

Boombox

Chris Savage
10-09-2006, 11:20 AM
Boombox,
·
·· Okay, now that I am sure of what you’re doing, my advice would be to test my original theory by measuring the voltage between ground and the top of R8 with your standard POT.· See what the voltage reads all the way through the adjustments and let us know.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support

Boombox
10-10-2006, 09:08 AM
http://forums.parallax.com/images/smilies/freaked.gif·Ok, I tried ramping the 10k pot up and down and I get .22 volts up to 12.2 with Vin set at 14.3. Any ideas?

Chris Savage
10-11-2006, 12:09 AM
Boombox,
·
·· That proves my initial theory, that you were trying to control too much voltage for the AD5220.· I don’t know if there is another digital pot that may work in this or not.·· The circuit may need to be revised to allow the functionality you’re looking for.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support

Boombox
10-11-2006, 12:40 AM
Can I do it with an op amp?

Chris Savage
10-11-2006, 01:27 AM
Boombox,
·
·· Yes, you can use an Op-Amp, but it’s not as simple as just sticking the op-amp in there.· The circuit needs to be configured properly for the ADJ terminal to get the proper current/voltage characteristics.· I would Google for some circuits.· I know I have seen a few posted that use an Op-Amp to drive the ADJ terminal, but a few extra parts are required, such as resistors and a transistor.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support

Boombox
10-11-2006, 01:53 AM
Ok Thanks! Do you have any other recomendatrions?

John Bond
10-13-2006, 09:59 PM
Hi Boombox

I am not an electronics expert so I·sometimes like reading the datasheet. (Yes, I know thats cheating) I have downloaded it from the link you provided and attached it. Read through it, specifically·the reference circuits on pages 9, 10, 11...

for example:-

An option·NI suggest is using a parallel binary input and 8 transistors. They also include a couple of examples of the use of opamps (though most of the opamps they specify are no longer available)

Herzogs suggestion of a negative voltage is also outlined here, specifically·-1.2V

Other suppliers have different circuits

About a year ago, I went the same·way you did but discovered that·both the digital pots I used only went to 5V or 5.5V. It's waiting on my bench for the serial digital option (0 to 24 in 255 steps or about 100mV a step).

Regards from Kwa Dukuza in Darkest Africa

John Bond