Strange problem with RS-232 Buffers

Hi,

I just spent the day being baffled by a strange RS-232 problem. I connected a barcode module which uses a MAX232A compatible RS-232 buffer from intersil. On my board uses a RS-232 buffer from Analog devices which has an enable pin for the receivers. The barcode module was sending data out without any problems when not connected to my board, but as soon as It was connected, I could get nothing on the scope. On the TTL side of the buffer the signal was visible, but on the RS-232 side nothing. After quite a few ours of frustration, I found that due to an etching problem on the PCB, the enable signal was at the wrong level, hence disabling the receivers. As soon as I enabled the receivers everything worked fine. It looks like the receivers, when disabled puts such a load on the other RS-232 output, that it kills the signal completely. Is this normal for RS-232 buffers ? Regards Anton Erasmus

Reply to
Anton Erasmus
Loading thread data ...

No, it's not.

[Also they're not buffers - they're transceivers.]

Another possible etch problem somewhere? Which device is it? (Clearly not a MAX232A-compatible device, since this doesn't have a receiver control line.)

Steve

formatting link

Reply to
Steve at fivetrees

Perhaps you should specify part numbers so we look up the parts?

I am using a Maxim 3243 which has an auto shutdown mode to save power. If it senses no input, it doesn't drive the output. Perhaps one of your tranceivers is doing something like that?

(Yes, I know this doesn't quite match your description, but it is all I can think of)

Alan Nishioka

Reply to
Alan Nishioka

It is a HIN232 device (MAX232A compatible device from intersil) driving into a ADM3222 device. The ADM3222 have a receiver enable input, and a transmitter shutdown input. The receiver input was driven high (inactive) due to a PCB fault. THE ADM3222 is a MAX232A equivalent in the sense that it uses similar circuitry with a charge pump using 100nF capacitors.

Regards Anton Erasmus

Reply to
Anton Erasmus

IT is an intersil HIN232 driving into an Analog Devices ADM3222 device.

The HIN232 seems to be a direct equivalent of the MAX232A. Replacing the HIN232 with a SIPEX SI232A did not alter the behaviour.

Regards Anton Erasmus

Reply to
Anton Erasmus

Are you sure that you are not driving the transmitter shutdown pin due to a wiring error and causing a "break" condition on that Tx line, that might cause some unexpected behaviour at the opposite side of the link ?

Are there any RS-232 handshaking lines connected between the systems and does the receive enable or transmit shutdown pins have an effect on these lines ? Does Rx enable/Tx shutdown pin alter the voltages on the charge pump capacitors ?

Paul

Reply to
Paul Keinanen

I checked both datasheets (briefly); didn't spot anything weird there. If disabling the receivers (which should cause the receiver *outputs*, i.e. TTL levels, to tristate) is affecting the RS232 line *inputs*, then you almost certainly have a wiring/tracking error.

Steve

formatting link

Reply to
Steve at fivetrees

The side where the transmission dissapears does not have a shutdown pin. It is a bought out barcode module.

There are RTS, CTS lines, but they are shorted together at the module.

Everything is working fine after I have enabled the Rx enable pin on my side. The problem was, that I was seeing nothing on the scope on the module transmit line. This lead me to believe that the problem somehow is on the module. Data was on the TTL side of the HIN232, but nothing on the RS-232 side. The charge pump voltage on the HIN232 is only +-5V, which rises to about +-6.5V when the Rx enable on the ADM3222 is true. I even replaced the HIN232 with a SI232A (Sipex), which have exactely the same behaviour. I am trying to understand the behaviour, before it bytes me in the arse at a later date. I would have thought that a RS-232 tranciever should be able to drive a signal at normal levels on a wire connected to a disabled RS-232 receiver on another device. A PC connected to the same ADM3222 have no difficulty in driving the signal lines when the Rx buffers are disabled. I have not checked which tranciever the PC has, but it is higly likely that it is a similar device using a chargepump to generate the RS-232 level voltages.

Regards Anton

Reply to
Anton Erasmus

'Nothing' as in : ? Active LOW ( Charge Pump V--) ? Active high ( Charge Pump V++) ? Grounded ( with what impedance ? ) ? Open Circuit ( ie can you pull the line easily with ~4K7 ) ? changes when you remove the ADM3222 load

Normally, yes. You do seem to have found another level of dependance. Some parts have Auto-power off, but that's not in the simpler devices, and those that do have it clearly stated in the data.

-jg

Reply to
Jim Granville

Indeed.

What happens into an open-circuit? Again, disabling a receiver should not affect the line.

Steve

formatting link

Reply to
Steve at fivetrees

You should do some current measures to narrow down the error.

- short circuit current from PC-TxD

- short circuit current from module-TxD

- current into ADM3222 with Rx enabled

- current into ADM3222 with Rx disabled

My guesses are:

- you have put in the ADM3222 the wrong way ;-(

- you burned the ADM3222 to some degree

- you mixed up receivers and transmitters on the ADM3222

BTW: Are you sure that you got the polarity on the receiver enable right?

/Jan-Hinnerk

Reply to
Jan-Hinnerk Dumjahn

It transmits fine into an open circuit. When I get time, I will investigate whether it is some strange layout problem. When tracing the connections on the PCB of the barcode module, the circuit used is the recommended circuit in the device datasheet.

Regards Anton Erasmus

Reply to
Anton Erasmus

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.