Re:KIO2/KPA100 Serial Comm Problems

[VR2BrettGraham]

N0SS continued:

>Well, I think I've found the culprit of my K2 serial comms problem... and
>it's the laptop!!!  Unfortunately!
>
>My IBM ThinkPad 380D shown the following voltages:
>
>Voltages:
>
>    TXD line (from K2PA100):
>       Quiescent:  -15VDC
>    Sending data:  +5VDC
>
>
>    RXD line (from IBM ThinkPad 380D):
>       Quiescent:  -5VDC
>    Sending Data:  +1.0VDC
>
>Given data provided by Jack Brindle, W6FB, the TXD line appears to be about
>1.4VDC under the minimum required voltage.

That is likely to remain flaky.

I can't find my copy of EIA/TIA-232-F, but ITU-T V.28 mentions the 3
volts as the minimum that I always remember.

5 volts may be what a driver should produce, as I did find some mention
of 2 volts allowance for noise at the far end.

Interesting - the usual culprit is something to do with a software
driver, as had been suggested.  Now is this some sort of intermittent
hardware fault, or could it be that one can't assume that an IBM
ThinkPad's RS-232 port(s) don't do real RS-232?  I was counting on
getting a ThinkPad as it's on of the few decent laptops that can still
be had with a real serial port!

Speaking of serial control of the K2, a few months ago I was looking
at the chirp (cricket sound when polled for frequency data by a
computer) & the data received by the K2 does get serious radiated by
the ribbon cable - though no amount of dressing of the cable made a
difference, suggesting rather than induced by the ribbon into
surrounding circuitry, it's induced into the lines in the ribbon itself
(power?).

Have a look with a 'scope probe around the ribbon & note the
difference in RXD & TXD-induced spikes coming from the ribbon.
Anybody have any idea why RS232-in-converted-to-TTL should
be worse than TTL-converted-into-RS232-out?

73, VR2BrettGraham

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Randy Rathbun]

It seems to me IBM does something weird with the printer ports too. I  
tried using my ThinkPad I got back in 2000 as a lap timer for slot  
cars and had all sorts of problems. Ended up just getting a cheap  
desktop and putting that on.

I have had really good luck with the Keyspan USB-RS-232 dongles though.

Randy Rathbun, NV0U
K2 #1981
[hidden email]


On Jun 13, 2005, at 7:56 PM, VR2BrettGraham wrote:

> N0SS continued:
>
>
>> Well, I think I've found the culprit of my K2 serial comms  
>> problem... and
>> it's the laptop!!!  Unfortunately!
>>
>> My IBM ThinkPad 380D shown the following voltages:
>>
>> Voltages:
>>
>>    TXD line (from K2PA100):
>>       Quiescent:  -15VDC
>>    Sending data:  +5VDC
>>
>>
>>    RXD line (from IBM ThinkPad 380D):
>>       Quiescent:  -5VDC
>>    Sending Data:  +1.0VDC
>>
>> Given data provided by Jack Brindle, W6FB, the TXD line appears to  
>> be about
>> 1.4VDC under the minimum required voltage.
>>
>
> That is likely to remain flaky.
>
> I can't find my copy of EIA/TIA-232-F, but ITU-T V.28 mentions the 3
> volts as the minimum that I always remember.
>
> 5 volts may be what a driver should produce, as I did find some  
> mention
> of 2 volts allowance for noise at the far end.
>
> Interesting - the usual culprit is something to do with a software
> driver, as had been suggested.  Now is this some sort of intermittent
> hardware fault, or could it be that one can't assume that an IBM
> ThinkPad's RS-232 port(s) don't do real RS-232?  I was counting on
> getting a ThinkPad as it's on of the few decent laptops that can still
> be had with a real serial port!
>
> Speaking of serial control of the K2, a few months ago I was looking
> at the chirp (cricket sound when polled for frequency data by a
> computer) & the data received by the K2 does get serious radiated by
> the ribbon cable - though no amount of dressing of the cable made a
> difference, suggesting rather than induced by the ribbon into
> surrounding circuitry, it's induced into the lines in the ribbon  
> itself
> (power?).
>
> Have a look with a 'scope probe around the ribbon & note the
> difference in RXD & TXD-induced spikes coming from the ribbon.
> Anybody have any idea why RS232-in-converted-to-TTL should
> be worse than TTL-converted-into-RS232-out?
>
> 73, VR2BrettGraham
>
> _______________________________________________
> Elecraft mailing list
> Post to: [hidden email]
> You must be a subscriber to post to the list.
> Subscriber Info (Addr. Change, sub, unsub etc.):
> http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/subscribers.htm
> Elecraft web page: http://www.elecraft.com
>

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Don Wilhelm-3]

> -----Original Message-----
> Have a look with a 'scope probe around the ribbon & note the
> difference in RXD & TXD-induced spikes coming from the ribbon.
> Anybody have any idea why RS232-in-converted-to-TTL should
> be worse than TTL-converted-into-RS232-out?
>
> 73, VR2BrettGraham
>

Brett,

Are you asking about spikes on the TTL level side or the RS-232 level side?
If it is on the TTL side, and the RS-232 side at the same time period has a
voltage beyond the 'region of uncertainty' between +3 volts and -3 volts,
one can only blame the receiving device, but I do believe the problem goes
further than that.

In the proper RS-232 world, the minimum levels (+/- 3 volts) are the minimum
that can/should be expected at the receiver end, while the driver voltage
can go anywhere greater than those values (up to +/-25 volts) and still meet
the spec.  The designer of the driving circuits have to make a decision
about how much cable they want to drive and what the safety margin they want
to use above the minimum.  Some laptops are limited on the power supply end
of things and may chose to drive the positive signal with only 5 volts but
will usually drive the negative with something like -12 or -15 volts.  They
lose some noise immunity on the positive side by doing that, but since in
RS-232 signalling the negative voltage is the logical 1 (or ON signal), they
can usually get away with it unless the noise source puts extra negative
going spikes on the cable.

I would tend to believe that something has failed in Tom's laptop rather
than thinking that it is a design flaw - yes, most laptops cannot drive the
200 ft. cables that used to be common in the RS-232 world, but that world
was usually not driven by laptop serial ports anyway.  In a typical home or
small office setup, the 5 volt positive level still provides more tha a volt
of noise imunity on the positive signalling level, and since the negative
signalling level is driven to a -12 volt level, there is plenty of margin
for noise on the level that will register a logical 1 at the receiving end.

73,
Don W3FPR
--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.323 / Virus Database: 267.6.9 - Release Date: 6/11/2005

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[VR2BrettGraham]

W3FPR asked:

>Are you asking about spikes on the TTL level side or the RS-232 level side?
>If it is on the TTL side, and the RS-232 side at the same time period has a
>voltage beyond the 'region of uncertainty' between +3 volts and -3 volts,
>one can only blame the receiving device, but I do believe the problem goes
>further than that.
>
>In the proper RS-232 world, the minimum levels (+/- 3 volts) are the minimum
>that can/should be expected at the receiver end, while the driver voltage
>can go anywhere greater than those values (up to +/-25 volts) and still meet
>the spec.  The designer of the driving circuits have to make a decision
>about how much cable they want to drive and what the safety margin they want
>to use above the minimum.  Some laptops are limited on the power supply end
>of things and may chose to drive the positive signal with only 5 volts but
>will usually drive the negative with something like -12 or -15 volts.  They
>lose some noise immunity on the positive side by doing that, but since in
>RS-232 signalling the negative voltage is the logical 1 (or ON signal), they
>can usually get away with it unless the noise source puts extra negative
>going spikes on the cable.
>
>I would tend to believe that something has failed in Tom's laptop rather
>than thinking that it is a design flaw - yes, most laptops cannot drive the
>200 ft. cables that used to be common in the RS-232 world, but that world
>was usually not driven by laptop serial ports anyway.  In a typical home or
>small office setup, the 5 volt positive level still provides more tha a volt
>of noise imunity on the positive signalling level, and since the negative
>signalling level is driven to a -12 volt level, there is plenty of margin
>for noise on the level that will register a logical 1 at the receiving end.

The spikes have nothing to do with N0SS's laptop - sorry for the
misunderstanding.  Tom's follow-up post reminded me of having
looked at this a while ago.

I hope N0SS has a dickie serial port & it isn't indicative of ThinkPads
in general (hard to get real serial ports on modern laptops & hope one
day to find another one with two & all the other goodies one day - have
some things to do that really can't be done any other way... will be
keenly watching for any final post from Tom).

The spikes I think are where the chirp (low level sound of cricket in
both receive & transmit audio - numerous reports here of it on receive)
when a K2 is being talked to.  Incoming data on KIO causes huge
spikes with 'scope probe in the vicinity of level far greater than spikes
seen when the K2 is sending data.

Haven't tried breaking out that line & shielding it, or doing something
to clean up the power to the interface chip - not sure why 232>TTL
would cause this & TTL>232 wouldn't.

73, VR2BrettGraham

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com