Serial port problem with Dell PC
Andyh
Posts: 13
Maybe you have heard of this before and or can recommend what to do:
I'm using a BS2p24 with 12-bit Adc to process signals from an op amp. Drives a 2x16 Parallel LCD display from the Plus Pack.· Some output pins drive LEDs. Not much more to the circuit at this stage.
My prototype pc board (not a Parallax demo board) uses a 4-pin SIP connector for a serial port connection for updating firmware, etc. I put the two 0.1uF capacitors on the pcboard per BS manual v.2.1 page 26, and made a cable with a DB-9 wired as page 26 on one end and the mate to the 4-pin SIP on the other.
When·the DB-9 made contact with·my Dell 4300S PC serial port the Dell immediately went ominously quiet; the 4-pin end was not connected. Upon boot-up the PC provided a message that it had shut down due to a thermal event.· Serious short seems evident, but my tech had ohmed out ALL the pins and I ohmed them independently.· We've tested the cable again and it's perfect.·
We cut the pin 6 (DSR) to 7 (RTS)·jumper, plugged it in, and it does not bring down the Dell.· Then plugged in the 4-pins on the other end and turned on the prototype board and the display did not come up so·whatever the Dell was putting on the four wires was not proper.· We took the board and cable to another PC and plugged it into the serial port and the display came up, so THAT computer was not screwing up the BS2p24.· So what's wrong with the Dell that would do this?
Keep in mind that this Dell computer has been used with the Parallax-supplied serial cable with the BS2p24 Demo board without a problem.· But I can't see any electrical difference between my proto with my cable and the bs2p24 demo board with its cable.
since the Parallax cable is straight thru I connected it in series with my cable and the display does not come up.· I unplugged my cable from the Parallax cable and the display comes up.· It's the Dell that's doing it ...
Thanks for any help.
Andy
I'm using a BS2p24 with 12-bit Adc to process signals from an op amp. Drives a 2x16 Parallel LCD display from the Plus Pack.· Some output pins drive LEDs. Not much more to the circuit at this stage.
My prototype pc board (not a Parallax demo board) uses a 4-pin SIP connector for a serial port connection for updating firmware, etc. I put the two 0.1uF capacitors on the pcboard per BS manual v.2.1 page 26, and made a cable with a DB-9 wired as page 26 on one end and the mate to the 4-pin SIP on the other.
When·the DB-9 made contact with·my Dell 4300S PC serial port the Dell immediately went ominously quiet; the 4-pin end was not connected. Upon boot-up the PC provided a message that it had shut down due to a thermal event.· Serious short seems evident, but my tech had ohmed out ALL the pins and I ohmed them independently.· We've tested the cable again and it's perfect.·
We cut the pin 6 (DSR) to 7 (RTS)·jumper, plugged it in, and it does not bring down the Dell.· Then plugged in the 4-pins on the other end and turned on the prototype board and the display did not come up so·whatever the Dell was putting on the four wires was not proper.· We took the board and cable to another PC and plugged it into the serial port and the display came up, so THAT computer was not screwing up the BS2p24.· So what's wrong with the Dell that would do this?
Keep in mind that this Dell computer has been used with the Parallax-supplied serial cable with the BS2p24 Demo board without a problem.· But I can't see any electrical difference between my proto with my cable and the bs2p24 demo board with its cable.
since the Parallax cable is straight thru I connected it in series with my cable and the display does not come up.· I unplugged my cable from the Parallax cable and the display comes up.· It's the Dell that's doing it ...
Thanks for any help.
Andy
Comments
·· You could really stand to post a few more details of your board and/or cable.· I myself did not get a good image of your connections from what you posted.· Out of curiousity though, if the Parallax Cable works, why use anything else?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
csavage@parallax.com
Second, I have managed to insert the wrong capacitor set-up [noparse][[/noparse]I did a mirror image of the proper one] because I was a bit tired and got confused when I built the set-up. In that case, the only thing that happened was no connections. I had to take the board apart and start over.
Third, you say that 'cutting the #6 to #7 line solved the problem. But, you also got this quite scary 'thermal event diagnostic'.
I would verify that you really had #6 to #7. You may have shorted something to ground by mis-reading the pins and cables. And, you also seem to have tested it on the second PC only after you cut the problem wires. [noparse][[/noparse]This is good as you didn't damage the second PC, but you really haven't fully analysed the problem.]
In conclusion, I really like building my own motherboards as the ones I buy are not only expensive, but not as compact or appropriate for my use. But, I now configure all the serial port pins [noparse][[/noparse]for programing] straight out - no capacitors. I have a plug in adapter that connects the board to my serial cable and the adapter has all the fancy stuff on it. In this way, my board is smaller and I no longer have to worry about being sure I get it right when I want to hand build another version. Also, I can connect those four pins to other serial interfaces [noparse][[/noparse]like some chip applications or another BasicStamp].
Sadly, you are on your own with these risks as you are building from scratch.
Anything you smoke, you replace at your own expense. The good side of this is you will probably be a better engineer because you really understand the costs of mistakes. We all kid about smoking ICs and boards, but it really hurts a bit.
When you get a problem like this, it is usually best to get a good night's sleep and then take another look at what you built. I continue to find that what I thought was perfect, was wrong. You tend to get in a rush to complete things. It is just normal human passion.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
G. Herzog in Taiwan
I thought I provided a lot of detail on the board and cable, but I did make you refer to the BS documentation.
I didn't put a DB9 on the proto board because if the schedule gets tight we might ship this to a customer, so it might as well reflect the final design. The final design would not have a DB9 because 1. It saves cost, 2. it shouldn't be routinely used - only to upgrade firmware, 3. I don't want someone plugging their computer in just because there is a standard connector on there. It happens. I'll put the DB9 on the product when I have a computer interface for customer data logging. 4. Etc.
Wiring Diagram:
Computer Cable
Cable Proto Board BS2p24
DB9M DB9F SIP-4 SIP-4 DIP-24
1 ...................... 1
2 ...................... 2.................. 1 ............... 1 ........................ 1
3 ...................... 3 ................. 2 ............... 2 ........................ 2
4 ...................... 4 ................. 3 ............... 3 ..] [noparse][[/noparse]................... 3
\...] [noparse][[/noparse]...
5 ...................... 5 ................. 4 ............... 4 ....................\... 4
6 ...................... 6 .....
7 ...................... 7 ....:
8 ...................... 8
9 ...................... 9
Kramer -
I thought the missing caps were another typo in BS manual v.2.0.
Re caps location: Three people have inspected the board - but the board was not connected to the cable during the 'thermal event'.
Re #6-#7: After cutting #6-#7 we plugged the cable into the Dell with no repeat of the thermal event. Jeff wired it and ohmed it and gave it to me. I ohmed it between each DB9 pin and each other pin, and to the SIP-4 and to all DB9 pins. Yesterday Guy ohmed it and said it was ok.
No confusion on #6, #7. I just took off the shell of the db9 and pins 1, 8 and 9 have no solder on them, pins 2,3,4, and five have wires on them (that agree with 1,2,3, and 4 of the SIP-4) and #6 and #7 have solder where the jumper was once attached. My eyes may be old but those are 6 and 7 right behind 1,2,3.
my only thought is a short to shell, but I routinely check that when I check pins on a custom cable and expect I did that. It's not there now.
I'm going to make an adaptor to put on the end of the Parallax cable. I still need another cable for the shop floor.
Re smoke: I've already smoked a processor ...
Andy
The capacitor symbol "\...] [noparse][[/noparse]..." should be above and to the left of "\... 4"
Lets try again, and defeat the forum software funny business:
Wiring Diagram:
Computer
Cable
Cable
Proto Board
BS2p24
DB9M
DB9F
SIP-4
SIP-4
DIP-24
1 ____________ 1
2 ____________ 2 __________ 1 _________ 1 ______________ 1
3 ____________ 3 __________ 2 _________ 2 ______________ 2
4 ____________ 4 __________ 3 _________ 3 __] [noparse][[/noparse]__________ 3
````````````````````````````````````````````\__] [noparse][[/noparse]__
5 ____________ 5 __________ 4 _________ 4 ____________\__ 4
6 ____________ 6 ___
7 ____________ 7 ___|
8 ____________ 8
9 ____________ 9
Andy
I have to assume your BS2 pin 4 is also tied to Vss, right? That's how the two grounds get tied together, so the signals are referenced to the same ground. Otherwise, you have floating grounds -- not a good thing. Still shouldn't cause anything to blow up when pins 6 and 7 are tied together, though.
Since pin 6 is an input, it is barely possible that pin has been shorted to ground inside your Dell (as a manufacturing flaw).· Then when you tie that pin to pin 7, the poor RS-232 driver tries to drive a dead short.· You should know however that RS-232 drivers are specified to be ABLE to drive a dead short without permanent damage -- it won't work, but it won't destroy anything.
If you just take a plain vanilla DB-9, tie pin 6 to pin 7, and connect it to your Dell, it sounds like you should get the same symptom.· And that's just wrong, there should be nothing about a serial port which can kill the computer for any reason, unless there's a manufacturing defect.
Post Edited (allanlane5) : 3/31/2005 8:25:26 PM GMT
It may have been a static electricity event. After all it is still colder and drier in America.· Taiwan always is humid, so static is less of a problem.· You may just have to inventory what is working and what is broken and move on.
I had a girl friend once that I fixed her brakes on her car and immediately the clutch went out.
As a result, she blamed me and never spoke to me again.
You really have to be careful of what you assume as cause and effect on an unrepeatable event.
This is a particularly ellusive one.
Life is like that.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
G. Herzog in Taiwan
Post Edited (Kramer) : 4/1/2005 7:52:10 AM GMT