wave for just flipped on me?

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

I wrote a piece of code to interface with an IC that latched on to a wave form. I just resurrected this project and it's not working. Looking at the data I see I was latching on to a low 50us dip. This 50us dip is now a rise? The only thing that changed is the wiring I use to connect to this IC.  So I'm wondering... Is it normal for an IC to revers its resting bias? Normally lo vs hi? Is there a common paradigm for this, like a pin that switches this behavior via a signal? For kicks I reversed the polarity but that didn't work nor did it hurt the IC.

 

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

there is the prof, I know I was not crazy.

 

This device does it's own clock. The wire you see with 4 pulses is the clock line (pretty obvious). The data line is the same polarity in both cases. So somehow I set the clock bias, but no idea how I would do that.

 

The colors are just old reference highlights, not relevant to this discussion.
 

Attachment(s): 

Last Edited: Sat. Aug 13, 2016 - 03:38 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I have no idea of what you're talking about! How about some context?

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

Kartman wrote:

I have no idea of what you're talking about! How about some context?

+1

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

You know you were not crazy?

 

Um.... Your presence here at AVR Freaks is solid evidence that you are.

 

Maybe you just discovered what was keeping it from working when you set it aside earlier. I've done that. "Well, no wonder I couldn't get it to work!"

If you don't know my whole story, keep your mouth shut.

If you know my whole story, you're an accomplice. Keep your mouth shut. 

Last Edited: Sat. Aug 13, 2016 - 01:01 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Slade had a song in the 70's "mama we're all crazy now".

you need to be a little crazy to work with electronics - getting them electrons to work the way you want is a challenge.

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

Did the image not show?

 

I cropped it out a bit. My point is that the clock signal is flipped. I'm reading a device that had a normally lo clock signal, and not it shows a normal hi clock signal.

 

____||||||___||||||_____|||||____

 

vs

____________________________

|||||      ||||||     ||||||     |||||||

 

The only thing that changed was the way the connected the device. The device has 4 digital out puts and 4 inputs.  The clock is one one of the outputs and data is on another. I played with all 4 inputs (hi / lo) but I still get this inverted clock signal. For reference this is a digital gameport device from back in the 90s.

 

 

Funny Torby, but it has been sold and used by many in its old form. The image I attached shows a capture of before and now. As I work with this I do now recall it did this in development last time. I remember working with it and thinking, I know that wave for just flipped but that was the last I saw it till now. There had to be one of these leads that flip it.

 

 

 

 

 

Attachment(s): 

Last Edited: Sat. Aug 13, 2016 - 01:51 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Well, the other way to invert the signal is to reverse your O'scope leads,

or have an Invert Signal set on the O'scope,

or have the scope set to subtract Signal 2 from signal 1, and signal 1 just happens to be sitting at ground, (unconnected),

 

Or the device is configurable, and it powered up in an alternative configuration,

 

For kicks I reversed the polarity but that didn't work nor did it hurt the IC.

What does that mean ? !

You reversed the O'scope leads?

You reversed the power supply leads? (Don't do that!)

You did something else?

 

JC

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

Well, the other way to invert the signal is to reverse your O'scope leads,

You can not reverse  one lead, you need two to do that. If you look at the inverted signal its on one wire? In my experience normally you have a situation where a device holds a line Hi via internal or external pull up. The other device then pulls this line lo to generate a clock. In this case it seems (odd to me) that the device I want to communicate with generates the clock on its own. So on the MCU my DDR is set to in, I stick an analyzer on it and you get the waveform shown.  That is all fine and good but this wave form has inverted itself. This is clearly shown in the image I attached.

 

If it does not make any sense for an IC to do this, that is ok, I can live with this. Though does no one here see what I'm talking about?

 

What does that mean ?

The power to the device.

 

 

 

(Don't do that!), I know that is why I said for kicks, I have plenty extra laying around.

Last Edited: Sat. Aug 13, 2016 - 02:37 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

After looking a bit more closely it is not flipped its just the normal resting point is now Hi instead of lo. I looked at the original board and I didn't use any pull down resisters and changing the port state on the atmega hi/lo does not effect it any. It will be near impossible to latch on to the start data without the first rising edge.

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

You mentioned that the device has four outputs but you are only monitoring two of them.  Could it be that the other two are complimentary versions of the first two?  i.e. a differential signalling scheme?

Letting the smoke out since 1978

 

 

 

 

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

You forgot to reverse the polarity of the neutron flow.

If you don't know my whole story, keep your mouth shut.

If you know my whole story, you're an accomplice. Keep your mouth shut. 

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

digitalDan,No I have all 4 on the logic analyzer. There is some activity on the two extra wires but it is not clear what it is. I do know that the extra two lines are a pass-true for another controller (daisy chain). And when I connect another controller I see just that. and what really irks me is the second controller is correct.

 

If controller A is connected line reset at hi. Connecting controller B, it rests at lo.

If I connect controller B directly it resets at Hi.

If controller B is connected first and controller A second, both reset hi.

 

This is so confusing.

 

Last Edited: Sat. Aug 13, 2016 - 06:17 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Ok so after playing with it for a bit it seem to change. Here is a screen shot of both resting hi next to a shot of both resting lo. Now to play the game of, what causes this. I hope its as simple as a short and not some crazy timing issue.

 

This is power via usb and to get this image I took a screen shot at the time both were normally lo, and without touching any wires I just reset the usb. Now they are normally hi.

 

Attachment(s): 

Last Edited: Sat. Aug 13, 2016 - 07:53 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You know, changing the scope trigger polarity can cause a signal to appear inverted!

 

Jim

 

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

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

right but

1) It's not inverted as I mentioned "After looking a bit more closely it is not flipped its just the normal resting point is now Hi instead of lo. "

2) It "flips" or whatever I should call it, on its own. Hi more often the Lo...

3) I can monitor two strobes at once and have seen where one is normal and the other is messed up. 

 

At this point I have all but given up and decided to write for both cases. Since the wave for is identical I don't have to change the main code. I just need to write to latch detection's.

 

Last Edited: Sun. Aug 14, 2016 - 12:36 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Is this an SPI device?

 

I have seen the same thing happen on a Texas DSP, the last bit determined the idle state and it did look confusing at first but alas it didn't make an apath of difference

 

I was also using a Salea Logger

 

Deja Vu!

 

Edit

 

And Hi Kartman and Cliff, have you been missing me lol

Last Edited: Sun. Aug 14, 2016 - 09:24 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

BigNoob, sure we've missed you. What have you been up to?

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

No it's just a normal single pin wave form. Interesting though... I still feel there is a way to set the resting bit.