USB-485, Low Voltage levels on 485A and 485B lines

[klieberschnitzel]

I've been having intermittent problems with communications using my NI USB-485 cable. 

So trying to determine if the cable itself was failing I probed the 485A and 485B lines on a scope.

I measured 2 different pulse peak-to-peak values on the 485A and 485B lines.

The 485A was generating 1.2Vpp pulses with a 625mV offset and the 485B was generating 3.09Vpp. 

Is this OK or is my cable failing?  I believe the latter and that it is time for a replacement. 

I did the same measurement using my PCMCIA-485/2 and both the 485A and B lines were a

strong 4.56Vpp level.

Any suggestions?  Do we need a new USB-485 cable?

Kathy Lieberman

[K C]

Hi,

When you measure the 485 lines you should measure between A and B. The level of A and B referred to ground should fall within the common mode range of the receiver.

If you are using a short (a few meters/yards) cable with just a point-to-point connection the cable is not that important. It will/should also work with two loose wires !!

What terminators do you use at both end ?

[klieberschnitzel]

Thanks for the quick response.

I'm measuring each line individually referenced to the Ground on the RS485 (pin 5 on a DB9).  Knowing the signals are differential I know the signals are of opposite polarity so I don't see a real problem with measuring them this way.  Maybe you can explain further.  Also, I successfully measured this way on a PCMCIA-485 card and got, as expected, a 4.5Vpp signal on both the 485A and 485B lines with opposite polarity.

In both cases, using the USB-485 cable and the PCMCIA-485, I had no termination other than the scope's input impedance, and I had no load on it when making the measurements.

I'm using a high input impedance scope.  

I don't understand why there is a differences in what I measured between the PCMCIA-485 card and the USB-485 cable, so I can only speculate, especially since the USB-485 cable is communicating intermittantly, that the unusual signals I'm measuring on the USB-485 cable suggests that it is failing.  Let me know if you agree.

Thanks,

Kathy

[K C]

Hi Kathy,

Well the (abs.max.) common range is normally -7  -  12V. Therefore, any voltage between these values referred to ground is in principle OK. To be on the safe side the value should be far from these maximum values.

You have to measure between A and B to see the ‘real’ signal that is at the input of the receiver. A difference of a few hundred mV is enough for the receiver.

However, more important is the termination. Specially if you do not use one.

Normally if I want to test a device, I do not care about the End-of-line terminator but I use USB to RS-485 device (FTDI 232 and a 485 transceiver) that needs to have an active terminator at one side. If I do not use this terminator, it does not work.

If you want to be sure things are always working use an active terminator at one side and a passive terminator at the other side.

(Need a drawing ?)

Kees

[klieberschnitzel]

I understand.

And, I would most enjoy a drawing on the active and passive termination you use.  This will also get me practicing transmission line theory.

Thanks,

Kathy

[K C]

Hi Kathy,

You will find the schematic of the terminators in the pdf. The values of the resistors are not critical. Just get back here if you have more questions.