Quickstart program download issue.

photomankc · 2012-04-03 13:45

Has there been, or are there any plans to, address the issue with the floating serial pin that seems to make it very difficult to download programs to the QS? I know I soldered in a pull-up on one of my boards and the "No Propeller Chip Found" issue went away when downloading programs. On my other board it is a fore-drawn conclusion that if I attempt to download a program it is going to fail multiple times, often needing a reset and sometimes needing me to remove and re-attach the USB cable. Any PC I use is the same. My old board was like that too till I added the pull-up.

Is there a fix planned for that or do we need to go on soldering in the resistors ourselves? I'm curious because I'm lazy and hate having to do it but it's not a huge thing. I can't imagine it's less than extremely frustrating though if you are a new user and can't figure out why it works now and then next time doesn't. If it's planned to fix it I may just hold off and wait to buy a couple I was planning on getting.

David Carrier · 2012-04-03 15:23

The transmit pin on the QuickStart is only a transmit pin when the program on the Propeller defines it as one, so it does float when it is not initialized. If your computer has an active serial connection to the QuickStart, but the QuickStart hasn't initialized the serial port, then the computer may receive undefined data. This is normal when connected to a device that hasn't initialized its serial port, but it shouldn't prevent the QuickStart from programming. There may be something wrong with your QuickStart or your computer. I would recommend contacting our technical support team to see if they can help you to get it working consistently.

— David Carrier
Parallax Inc.

Rayman · 2012-04-03 16:47

It's hard to imagine a pull-up resistor there having this effect, but I remember this coming up in another thread.
photomankc seems pretty convinced, so maybe there is something there...
Only thing I can possibly see is that incoming random data is keeping an interrupt busy and delaying the Prop Tool somehow...
Or maybe the Prop Tool doesn't flush the receive buffer well enough...

I had my own Quickstart issues on one computer, but ever since I rebooted, it went away...

jazzed · 2012-04-03 16:58

The loader gets noise between reset and the Propeller ID sequence.
The loader then interprets the noise as a protocol failure.
All we can do is get the timing right with the loader or add a pull-up.
I hear the latest Propeller Tool has a parameter for setting the post-reset delay.

Ariba · 2012-04-03 17:42

The PropBOE has now this pullup at the TX pin, so also Parallax seems to know about this problem.
I think it depends a lot on the code stored in EEPROM, if the TX pin (P30) is set as high output, then the logic level at the buffer input will remain high between a reset and the begin of the download. So writing the following code into the Quickstart EEPROM may have the same effect as a pullup:

PUB HighTX
  outa[30] := 1
  dira[30] := 1
  repeat
    waitcnt(0)

Andy

Rayman · 2012-04-03 17:46

Interesting idea Ariba... So, just having the PST or fullduplexserial objects started so do this too, right?

Ariba · 2012-04-03 18:43

Yes, I think so. As long as the reset not occures when you send a zero bit over serial out, but the chance is not very high (depends on how often you send something).

Andy

Rayman · 2012-04-03 18:55

It would be very ironic if this turned out to be true...
Gone from a situation where having a serial object started caused problems to a situation where the serial object solves the problem...

photomankc · 2012-04-03 22:27

I can only say that running PST doesn't seem to improve anything, that's exactly what I have been using in all my recent projects. Also, just like the other board if I plug a 10k from 3.3 to TX then the issue goes away. If it is not there then there is some random chance that it will fail to download. Tonight it was much less frequent than this afternoons experiance. I'll try giving the tech support line a call. Not trying to beat up on the product, I really like it. Just a bit annoying when it starts this.

Leon · 2012-04-04 00:58

I don't have any problems with my QS boards. It looks as though yours has an intermittent fault.

MagIO2 · 2012-04-04 01:35

Hmmm .... have to try this out this evening ... when I started with the QS I also had no problems, but somewhen the behaviour explained above also was true for my QS.

If I switch to PST and run a program which is doing continuous serial out, switch back to PropellerTool and then write to RAM chances are best.

I think what has been changed is, that I switched to a newer version of the PropellerTool. The older version did not have this problem!

photomankc · 2012-04-04 22:28

Well I am using the latest PropTool and I just got another QS from Gadget Gangster and what do you know, same issue. Then I remembered something about a setting for the delay in the PropTool and changing that helping sometimes. Found that and tried making it longer. It got worse. More time seemed to make the download failure happen more often. Then I made it little shorter and so far the problem is gone. I haven't been able to reproduce it any more with the delay set to 70ms. At 90ms it was so-so; failed now and then and at 100ms I pretty well could not download a program at all. We'll see if that's the end of it. I just can't see how three boards could all have the same fluke issue so it's got to be something about the Q/S, my PC, and the PropTool or all three but it's 3 for 3 thus far. If this is the fix then that's super! No soldering resistors needed.

jmg · 2012-04-17 17:27

photomankc wrote: »

Well I am using the latest PropTool and I just got another QS from Gadget Gangster and what do you know, same issue. Then I remembered something about a setting for the delay in the PropTool and changing that helping sometimes. Found that and tried making it longer. It got worse. More time seemed to make the download failure happen more often. Then I made it little shorter and so far the problem is gone. I haven't been able to reproduce it any more with the delay set to 70ms. At 90ms it was so-so; failed now and then and at 100ms I pretty well could not download a program at all. We'll see if that's the end of it. I just can't see how three boards could all have the same fluke issue so it's got to be something about the Q/S, my PC, and the PropTool or all three but it's 3 for 3 thus far. If this is the fix then that's super! No soldering resistors needed.

This sounds predictable, and even expected, given this boot order ?

PropBoot wrote:

2. The Boot Loader performs one or more of the following tasks, in order:
a. Detects communication from a host, such as a PC, on pins P30 and P31. If communication from a host is detected, the Boot Loader converses with the host to identify the Propeller chip and possibly download a program into global RAM and optionally into an external 32 KB EEPROM.
b. If no host communication was detected, the Boot Loader looks for an external 32 KB EEPROM on pins P28 and P29. If an EEPROM is detected, the entire 32 KB data image is loaded into the Propeller chip’s global RAM

So a PC-link-pin that is totally FLOATING is a very bad idea, as the boot ROM take a peek at that during load, and that also means EE patches will not help ?
Sometimes you just need a resistor

cavelamb · 2012-04-17 21:42

Leon wrote: »

I don't have any problems with my QS boards. It looks as though yours has an intermittent fault.

+1 No problems (so far)

Quickstart program download issue.

Comments