Hi,
Of late I have been having issues with interfacing serial "RS232" sensors to controls ports on the loggers. Most of the issue appears to be the widespread use of the Maxim MAX232 IC to drive the RS232 ports of these sensors. The MAX232 requires the TX data line to be held high for correct operation and as far as I an tell this is not the case on the control ports. Typically I prefer to poll sensors for data rather than let them free flow into the logger. If anyone has any low cost (I know that SDM-SIO1 works with these sensors) workarounds I'd appreciate some input.
Thanks,
IslandMan
Hi IslandMan,
Are you using your own MAX232 to change the TTL levels of the control ports of the logger to RS232 specs prior to interfacing to the external sensors MAX232 interface?
As you probably know, the control ports don't actually generate "true" RS232 levels, only TTL levels. Specifying LoggerNet serial instructions to output RS232 on the control ports is some what of a misnomer as I understand it. It should take care of the idling state of the TX pin, but doesn't generate true RS232 levels. When in doubt, use your own MAX232 which is designed to translate TTL levels to RS232. Then send the RS232 signals to the sensors MAX232. Then in the serial out instruction you would specify TTL levels. Hope this makes some sense,
Steve
Hi Steve,
No, I am not using a MAX232 on the logger I am referring to sensors with RS232 outputs that utilize the MAX232 chip.
I am aware that the "RS232" on the control ports isn't RS232 in the true sense of the specification but relies on certain RS232 circuits ability to function with TTL logic. Case in point is the old Tandy computer some of us old folks used with CR10 loggers directly back in the day.
You might think that with a sensor that streams RS232 you would be able to just connect the RX line and a ground t receive data. This is not the case with a sensor that employs the MAX232 chip, TX needs to be connected and it needs to be at the positive RS232 logic level idle state to function at all.
I have encountered this with a few sensors I have worked with recently, currently it's the PTB210 from Vaisala. I have an email into Vaisala to confirm the RS232 output circuitry. I'm looking for a lower cost (than SDM-SIO1) fix to make this work somehow.
Thanks,
Dave
I think you will find the problem with the Maxim chipsets is not due to the signal levels per se, which are acceptable to the chipsets (we have our own sensors and know many others that work). It is usually down to some sensor suppliers using the auto-shutdown feature many Maxium chips support which turns off the serial port if no valid RS232 level are detected on the inputs. By a valid signal they are looking for the output idling low at a negative voltage, rather than sitting at zero volts between measurements as the control ports will.
IF you are lucky this feature can normally be turned off in many devices and they will then work OK.
BTW we have used the Vaisala PTB220 with the control ports of the loggers without issue. The PTB210 is likely different especially as they offer a version with RS232/TTL as an option.
Hi Andrew,
I too have successfully interfaced the PTB220. If I put the PTB210 in the constant output mode and connect the TX line to 5V the sensor outputs fine. I'm not sure I can have it come in this state permanently. I ran out of time to test it, in Arizona this week. I have a query into Vaisala as well but haven't heard back. I was hoping there was a quick solution as I have another on-going project that has a number of these "RS232" sensors and can't afford add the cost of 4 SDM-SIO4 modules to the project. Thanks for your input,
Dave
An interesting article to read on Maxim Shutdown and RS232 "compatability" modes can be read here:
http://www.maxim-ic.com/appnotes.cfm/an_pk/882
A couple of points you might note from that are:
a) if your problem is caused by the sensor shutting off its serial port due to having the Autoshown feature, which cannot be disabled in software or by a simple hardware mod, if the sensor has input lines, you can usually prevent the shutdown by tying a free input, e.g. a handshake line, to +5V as this will trick the sensor into thinking an active RS232 device is connected. This seems to be the case for the PTB210 from your findings.
b) some Maxim parts only guarantee an output of +/-3.7V as RS232 signals. These are called compatible rather than compliant. As the minimum "high" level input for the CR1000 control ports is 3.8V you are on the edge of this working. In practise it will normally work as the control port has a much higher input resistance that an RS232 port so won't pull down the output driver much, but if the cables are long, etc, etc beware.
* Last updated by: aps on 11/19/2009 @ 4:16 PM *
Hi Andrew,
I had looked at the app sheets for the Max products but I haven't heard from Vaisala yet so I don't know what part I'm dealing with. Could just be a transistor setup like some other sensors I've interfaced with and not the Max chip at all. I hesitate taking the sensor apart to look. Once I hear back from Vaisala I'll update the post. Thanks for the help and insight.
Regards,
Dave
* Last updated by: IslandMan on 11/22/2009 @ 4:31 AM *