Strange Prop issue when booting from EEPROM..

markaeric · 2011-06-23 09:38

Hey guys,

I'm currently using DIP40 Propeller Education Kit and am having an issue with my program when it's loaded from the EEPROM. I can tell it's partially working by some 7-segment LEDs and a few other things. However, if I load the program directly to RAM, just about everything else works fine.

I tried using another Prop, but the problem remained. Unfortunately, I don't have an extra DIP eeprom, so I can't try that just yet.

I'm using BST on Ubuntu, and I don't get any errors no matter which way I compile. I haven't had an issue with this before.

One last thing, this doesn't appear to be related to the PropPlug reset issue.

Any thoughts on what might be the issue would be greatly appreciated.

Thanks,
Mark

Leon · 2011-06-23 09:45

Try removing and replacing the EEPROM, there might be a poor connection. If that doesn't help replace it.

markaeric · 2011-06-23 10:12

Thanks Leon. I tried reinserting the eeprom, and some other stuff, but it's still a no-go. Wouldn't an eeprom issue be identified during the write verification?

Phil Pilgrim (PhiPi) · 2011-06-23 10:31

markaeric wrote:

One last thing, this doesn't appear to be related to the PropPlug reset issue.

Are you sure? What's the condition of the Prop Plug and its connection to your computer when the problem occurs?

And, yes, an EEPROM issue would be identified during verification.

-Phil

Duane Degn · 2011-06-23 10:43

I'm assuming you're using a breadboarded circuit right? Are you using a good gauge of wire for a breadboard (22 or 24). Breadboards can have bad connections even with good wire. If you're using wire that is too thin it will make things even worse.

You might want to try switching out the wire you have now.

markaeric · 2011-06-23 10:58

Thanks for the replies.

Phil,

The issue persists whether I have the prop plug connected to the programming headers or not.

Duane,

I am using a breadboard. The thing is, I've never altered that portion of the circuit on the breadboard (using the reference layout). The rest of the circuit (save for a few things) works fine as long as I load the program to ram. And since BST doesn't report any problems when programming the eeprom, I have to imagine the connections to it are fine. But who knows.

-Mark

Phil Pilgrim (PhiPi) · 2011-06-23 11:02

The only thing I can think of is that the EEPROM verify routine reads the data at a slower rate than the EEPROM boot routine. If that's the case, you might want to make sure you have bypass caps on both the Propeller chip and the EEPROM. Also, double check to make sure there's a pull-up on the EEPROM's SDA line. (If it's 10K, it wouldn't hurt to reduce it to 4.7K.)

-Phil

markaeric · 2011-06-23 11:41

Phil,

I don't have any bypass caps in the circuit.. I'll go ahead and try slapping some on. The pull-up on the eeprom is indeed 10k - I'll try it with a 4.7k.

On another note, I placed a bunch debug comments in the spin cog that appears to have the issue, and have found it to get stuck at the following code:

  ifnot (LastBtn == Btn) 'if there has been a change in button state since the last go around
    Case Btn

      0:
        Button := LastBtn := 0
        BtnTimer := 0

      BtnSel:
        BtnTimer := cnt + SelBtnHld
        Button := LastBtn := Btn

      BtnSav:
        LastBtn := Btn
        SaveSettings

      BtnCfg:
        LastBtn := Btn
        CfgMode++

      Other:
        Button := LastBtn := Btn

I just don't get it.

markaeric · 2011-06-24 07:43

Still can't figure this out. I've added bypass caps to the Prop and eeprom, as well as the change to the pull up resistor. The thing that really gets me is that I tried using a program that's about 80% different save for some of the code I posted above. It too only sees an issue when loaded to and run from eeprom. I'm starting to doubt that it's a hardware issue.

markaeric · 2011-06-24 07:51

D'OH! I just found that my "Save" button, which I don't currently have wired up, was set to be connected to pin 28! This has to be the issue..

Edit: Yep, that was it!

Thanks for all the suggestions everyone. If anyone needs me, I'll be walking down the Hall of Shame.

Grrrr...... Can't figure out how to change the thread's status to "Solved"

Duane Degn · 2011-06-24 08:37

Don't be too hard on yourself. We've all done something like this (at least I know I have).

To change to "Solved," edit your first post, "go advanced" and you should have access to the little pull down menu with "Unsolved" and "Solved."

Duane

Strange Prop issue when booting from EEPROM..

Comments