LM7815 power supply question

[nickb]

Getting there. Got the jack wired up, worked up the nerve and started up the board. Powers up fine, everything seems to work. The LED's on the board do the right thing. And the kit comes with a MCU preprogrammed with a cheesey program that makes moving light patterns on LED's by button presses. So, at least I know the inputs and outputs work, the power supply on the board is OK. and the power supply I made at least works.

When I have the board powered for a long time (hour?) the heat sink does get hotter than I expected. Not too hot to touch, but ... hot. The manual said 10 to 15v, so I used a 12v regulator.

Still can't get the AVR studio to see the board. I got a few really cheap USB-To-RS232 cable/converters, that came with drivers. Ive been installing and uninstalling drivers for a few days now, and I think Windows XP sees the cable, but either the cable/converter is no good, or maybe the RS232 port on the board is burned out (I REALLY hope that's not the case)

The last stretch, I opened a RS232 communications program I downloaded, and could send characters in, and when I hit return I got some garbage back. So, I think that means data is going all the way into the cable and to the board, and the board is responding although the comm's might be all screwed up. When I disconnect the board end of the cable, I dont get any response when I send garbage char's in.

So far, haven't found any way to test the RS232 on the board other than get a good working connection. so stuck for now.

I don't think it's surprising at all that it gets hot. How much current does it actually draw?

If you can send data but get garbage back, check the RS232 settings in in the dumb terminal your are using. If you get over that hurdle you and connect using AVR Studio ( it's 20 years since I last used it but I don't recall RS232, just their own dedicated ISP ) look for the RS232 settings in there. If there are none you can set them in windows Device Manager -> Com ports -> Properties->Port Settings

[mikepukmel]

You lost me, Mike. What is it you're building? I must have missed something. I see a thread about regulation and then the last post about RS232, etc. Curiosity is getting the best of me. Is there maybe an associated thread I missed?

Sorry Dude, I forget that this is a blog thread, that has a question, or topic and when that's resolved I should start another thread. It was a ham klutzy segue into what should be another topic.

So, the power supply works fine. Thanks to everyone for the help. The PS is powering an Atmel STK500 dev board, for programming Atmel microcontrollers (goal is to learn something about programming digital signal processing for audio).

Board powers up, initializes, the LED's do the right thing. The board comes with a chip that has a cheesy program on it, that responds to button presses by making a moving light pattern on the on board LED's, so I know that the chip socket is powered, and the i/o work fine.

Next step was to hook up a RS-232 to USB connection to my old PC, fire up Atmel software studio and see if it can talk to the STK500 board.
That connection is the problem.

I'll stop posting stuff about the RS232 on this thread and wind up the power supply questions. Thanks.

[mikepukmel]

I don't think it's surprising at all that it gets hot. How much current does it actually draw?

If you can send data but get garbage back, check the RS232 settings in in the dumb terminal your are using. If you get over that hurdle you and connect using AVR Studio ( it's 20 years since I last used it but I don't recall RS232, just their own dedicated ISP ) look for the RS232 settings in there. If there are none you can set them in windows Device Manager -> Com ports -> Properties->Port Settings

Thanks Nick. I have a pretty good Fluke (not high end) but should handle the current measurement. have to figure out how to wire that up. Im not sure, but the board folks said "min 500ma" so Im guessing around there. I will work up the courage to measure the current tonight.

Thanks for the tips on RS232, its been ages. Just hooked up a cable, shorted pins 2-3, and get a round trip. So I think (maybe) the cable is OK. Hoping not since if its good, that might mean the receiver chip on the board is toast. But I will try fiddling with the data settings. The Atmel software doesn't have any settings but I think there are all in windows.

The 'better' quality cable with driver came in today. Should pretty much tell me if the RS232 chip is toast. Even on this really old board, its all surface mount IC's so a bit beyond my skill set to replace any of it.

[mikepukmel]

Wahoo moment. Got the new cable, and 'better' driver, installed, didn't work. So I went on a long path of installing older and older versions of the AVR Studio software. Finally, one version installed, and when I tried to connect, got a message "The firmware is version 1.x, install version 2.x?" I almost fell over. Followed the instructions, the firmware upgrade succeeded, then the firmware validation. Then, went back and forth with newer version of AVR Studio. They all seem to connect and 'see' the board just fine now. So, it might have been that the firmware was very, very old, like experimental. I think the guy bought this dev kit, put it in a closet and never touched it. Also, I plugged in and unplugged ... repeatedly many times, so maybe the little sockets on the female DB9 on the board had some 'crud and corruption' in there (thanks to Mr Cassavant, my 11th grade metal shot teacher for the phrase) that worked its way out. Im a gonna be a DSP programming fool, until I fry the chip.

[mikepukmel]

You lost me, Mike. What is it you're building? I must have missed something. I see a thread about regulation and then the last post about RS232, etc. Curiosity is getting the best of me. Is there maybe an associated thread I missed?

Hi Dude, sorry, I missed your question. I got a Atmel STK500 development board. its got a programmer on the board for their microcontrollers. But its an ancient design, so the port to the programmer is RS232. The power supply is for the board, which has a power supply of sorts. I don't understand the details very well, but the board has diodes and a regulator, but the mfr says the board should be 10 - 15vdc. So, the power supply will power up the board, which has some chips used to program the actual microcontroller under test. The nice thing about this old board is that it has a whole bunch of different sockets for many of their chips that come in PDIP. The board also has io ports that you can connect up either off the dev board to a breadboard, or to onboard set of buttons and LEDs. The board kit comes with a really old (outdated) microprocessor that you can plug in, that has a cheesy program that lights the LED's on the board in some pattern, to verify that the board is running OK.

There are a bunch of lectures online from the EE department at Cornell, for a course that teaches microcontrollers. That's the only reason I got this kit, is that I can go through the lectures, and also the homeworks and lab projects and build the stuff and learn how it works. They have one lab project that is DSP audio synth. Im interested in the other stuff, but really interested in audio and DSP. Not like Im going to be designing any DSP/Guitar boxes, but its something ive always been interested in since one of teh kids in my grade school picked up a real Moog mini synth back in the 70's at an accidental garage sale.

So, wanted a couple of power supplies, one for the board and that could power the chip running whatever program, and another one or two to power other 'stuff'. One of the other lab projects uses a 2 x 16 (I think) LCD display, and a keypad. They go over how to debounce the key buttton presses and send text and numbers to the LCD.