AVRISP clone from Digikits unreliable after a few months

Go To Last Post
18 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello there,
in february 2008 I bought an STK500v2 AVRISP clone from Digikits, and I used it with pleasure on OS X for a long while. All I had to do was install the CP2102 USB-Serial drivers, and avrdude was happy.
I had to give up electronics for a few months, and I finally got back to it today. I had a breadboard with an unfinished project I didn't really care anymore, so I decided to reflash the fuses of the mega8 that was on there so it wouldn't need a crystal, and move to the tiny45 I wanted to use for a newer project.
Well, though luck: avrdude just couldn't read the signature of the device (it reported 0xffffffff). Everything was fine, and I even completely dismantled the breadboard and re-mounted only the mega8 and a 78l05 for the process. Nothing.
I upgraded avrdude, redownloaded the usb-serial drivers and reinstalled them, and so on. Still nothing.
A couple of kernel panics later (!), it finally came back to life... sort of. Essentially, it works once every five times or so.

For instance, I tried having it check for the mega8, and I got this:

octavarium:~ jollino$ !avr
avrdude -pm8

avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.01s

avrdude: Device signature = 0x1e9307

avrdude: safemode: Fuses OK

avrdude done.  Thank you.

Immediately after that, I tried again and I got this:

octavarium:~ jollino$ avrdude -pm8

avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.01s

avrdude: Device signature = 0x1e93ff
avrdude: Expected signature for ATMEGA8 is 1E 93 07
         Double check chip, or use -F to override this check.

avrdude done.  Thank you.

I honestly don't know what else to check. How likely is it that a programmer just dies out of not being usage? Might it be some dust that got in the breadboard or the programmer? (They've been sitting on a remote shelf rather than inside a drawer or a box, my mistake.)

I mean, it's not a big deal if I have to get a new programmer. There are some cheap ones on ebay that I can get for less than €30, so if I need to get a new one, so be it.
I wonder if there is something else I could try before shelling out money on a new AVRISP clone, though. I'd rather spend that on new parts. :)

Thanks!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

so you are going to buy another clone/cheap-o?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

@stevech: considering I may go through another "halt" phase soon, I'd rather avoid buying an official programmer that costs at least twice as much. :)
@SteveN: if the problem comes up again, I'll try; however, after I posted I tried cleaning the USB connectors on the cable with some alcohol (they looked slightly oxidized) and connected it to another USB port. It seems to be working fine now... Definitely weird.
I'll get a couple extra ribbon cables anyway soon.

By the way, is there any way to crimp IDC connectors on ribbon cables without resorting to specific pliers?

Thanks!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

jollino wrote:

By the way, is there any way to crimp IDC connectors on ribbon cables without resorting to specific pliers?

Thanks!

I'm using a plain vice

/Bingo

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Bingo,

What do you use to crack walnuts?

Jollino,

Use a Mole grip. Adjust it to the exact closure size, and your IDC is made perfectly.

David.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

That's a great idea, thank you!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I learned a long time ago that you buy good tools once and bad tools twice.

Cheers,

Tom

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Is a bad tool at half the price really a half price tool? Not if you have to buy it twice and/or have to spend hours debugging why it isn't working.

Buy it once, buy it right, you save in the long run in more ways than just your pocket book.

Writing code is like having sex.... make one little mistake, and you're supporting it for life.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

jollino wrote:
@stevech:
By the way, is there any way to crimp IDC connectors on ribbon cables without resorting to specific pliers?

Thanks!

i use my feed, insert the cables to IDC connectors, put it on the floor and than step on it.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

david.prentice wrote:
Bingo,

What do you use to crack walnuts?
David.

The same as for inserting a 40-pin DIP in a socket .... This one :-)

/Bingo

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

david.prentice wrote:
Bingo,

What do you use to crack walnuts?

Jollino,

Use a Mole grip. Adjust it to the exact closure size, and your IDC is made perfectly.

David.

There is nothing wrong with using a vice. It is a tried and trusted procedure. The jaws are nicely aligned and apply equal presure over the whole IDC. E.g. try to apply a 40 pin IDC connector and you quickly learn the advantage of the vice over the mole grip. Like, for example, no broken or mangled connector housings.

Put two pieces of wood in front of the vice jaws if you are sheepish.

Officially, of course, you use IDC pliers / crimper from $30 upwards to $100.

Stealing Proteus doesn't make you an engineer.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I would guess that
more people have available feet
than people have available pliers
than people have available mole-grip
than people have available vice
than people have available proper IDC tool

The advantage of mole-grip is that you can set the closing action to the exact size, and you have a "feel".
Aluminium or plastic jaw covers are advisable for a vice. And I am sure you can mark the angle of the handle that corresponds to the exact closure size, or possibly set a jig to limit the jaw or handle travel.

Personally I would not recommend method #1 or #2.

On a related subject, crimping individual pins for Molex or 2.54 x 2.54mm housings are very fiddly. I use a £30 RS universal hand tool. You have to align everything very carefully. The professional tool is over £100.

David.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The programmer now seems to work most of the time, but ultimately the AVRs won't. I tried a simple

int main(void)
{
	DDRB = 0xff;
	PORTB = 0xff;

	while(1);
}

on two different ATtiny45s (one of which had never been used in the past) with a LED on PB0 (sourcing), but it just won't come up. It looks very very faint, as if the resistor were in the order of tens of thousands of ohms, but then again it does that even if the chip is erased (but won't if I disconnect the programmer from the board). Needless to say, the resistor is the typical 470 ohm I usually use for LEDs.

What's surprising is that when it works, it manages to verify successfully. Could anyone confirm it is indeed the programmer?

Also... at this point I'll go for the AVRISP2. I don't need the STK500 since I'm more at ease working on a breadboard anyway.

I'm actually a bit confused about what's happening, though. The 78L05 seems to output only 4.10 V (even with no load whatsoever). Note that the power supply outputs about 9.80 V, and that should be more than enough for the 78L05 to work properly.

May it just be the breadboard that's not as reliable as it was a few months ago? (I'm not sure if that makes any sense, but please bear with me... you can tell I'm still a big newbie, can't you?)

[Edit: I did try several 78L05s, some old and some new.]

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

A 78L05 is the TO92 packaged version isn't it. The free air dissipation is of the order of 300mW or output current of 60mA. And it will be VERY hot.

You have obviously got a problem, and it is most likely to be with your wiring. Bingo may be abe to help you with a few tools.

However if you have not fried your 78L05, I suspect that you do not have the capacitors recommended on page 9 of the LM78Lxx data sheet. And your device is busy oscillating. Hold an AM transistor radio next to it.

David.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yes, the 78L05 is TO92. I did put the capacitors required (I learned that the hard way a long while ago :)), but I also tried a TO220 version (7805CZ? I don't remember exactly right now) I had lying around and it's the same; that one's output is rated at 2 amps, definitely more than I'd conceivably need for a couple of LEDs and a small mcu.

I defnitely do not rule out that my wiring has some issues; I haven't done this in a long while and I feel like I'm learning again. However I'm wondering whether dust in the breadboard might play a role. After all, I have had pots misbehave because of dust, so who knows. Maybe some compressed air might help.

Alas, I have no AM radios around here. :)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Scope will see anything the AM radio can hear.

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I would guess that jollino's budget does not quite stretch to an oscilloscope.

Although I have used FM radio for the last 30 years, every radio that I possess will receive AM (except for one DAB set). I would have thought most households are more likely to have an AM radio than a scope.

David.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

An oscilloscope is definitely one of the things that I put on my "vorrei ma non posso" list (I wish but I can't). The main problem is that I'm definitely a hobbyist who has to take his time between electronics sessions, so spending too much money on what I might end up using once in a long while is something I try to avoid, especially right now. :)

Anyway, after close brain inspection, I remembered that this power supply used to deliver higher voltage than labeled: when set to 6 V, it would be about 9 V; when set to 9 V, it would be about 13 V; when set to 12 V, it maxed out at almost 18 V. Now it's quite the opposite.

Long story short, I'm running the blinking thing on 4 AA cells (which curiously enough seem to work with the 78L05, albeit it only delivers 4.80 V) and it works just fine.

I probably blew the power supply when I tried it to see if a satellite receiver was dead; the original supply delivered 2 amps, while this one pushes barely 500 mA. I hadn't checked, so it probably overheated — or something — and it died.