# USB to serial 232 adapter for GE SNP w/LM90

R

#### raywins

I have a 1 year old HP laptop with no 9 pin COM1. I have purchased an "I/O Gear" brand USB/232 adapter, model GUC232A. It's installed properly (after talking to their tech support}. I have a working SNP adapter between it and a Series 90 Micro.

I'm running LM90 v9.02 under WinXP Home Edition. When I switch from "offline" to "monitor" I get LED indications on the USB/232 adapter that LM90 is trying to talk to the PLC, but LM90 just displays it's "no communications" message at the bottom of the screen.

Connecting to the Micro through the SNP cable from COM1 of my tower running LM90 under Win2K works ok.

Anybody have experience with USB/serial 232 adapters and GEF SNP? Like maybe something that works?

H

#### Hank Nusko

The problem is that the converter needs 5V that is supplied by the standard comm ports but not supplied by the USB serial converters. The 5V needs to be added from either an outside source or borrowed from another port that provides it. Hope this helps

M

#### Mark Hensley (Kepware)

Hi Ray,

One thing that may be causing a problem here is the BREAK character. To initiate an SNP communications session the PLC expects to see a break character come across the serial interface and be held in the break state for a period of around 400milliseconds(by spec). A BREAK character is something the UART(the chip that does serial communication) generates. It does this by driving the serial line to single state without any start or stop bits.

Normally in systems where the native serial port is being used, the standard Windows Comm Library is driving the UART directly and can easily instruct the UART to enter the BREAK state. Thus allowing the GE PLC to see the BREAK condition and prepares itself for what is called an "Attach Message".

If the USB serial port does not send this break character through or does not hold the break condition long enough you will never be able to talk to your PLC. Technically you may find that this is the case. The reason I say that is based on the fact that few modern products these days use the break condition. As such, vendors of products like your USB serial adapter do not even think to support the transmission of the BREAK character. You may want to try using a Digi Internation product. They have USB serial adapaters but more important they specifically play in the industrial space and as such, are aware of the needs of industrial products like your GE PLC.

Hope this helps.

L

#### Lynn at Alist

Just an educated guess - SNP requires a 'long break', which is an older UART concept where the data line is pulled artifically low for several charector times - a kind of a controlled 'framing error'. Very likely your USB/232 product either cannot do a long-break, or is trying to simulate it in a manor the GE PLC doesn't accept.

I cannot offer any advice on USB/232 that support 'long break'. Perhaps you can set the PLC to use SNPX (or "breakless SNP). This might solve the problem.

Best Regards

Lynn August Linse
IA Firmware Specialist, Digi Int'l (www.digi.com)

A

#### Alex Pavloff

Lynn at Alist wrote:
> Just an educated guess - SNP requires a 'long break', which is an older
UART
> concept where the data line is pulled artifically low for several
charector
> times - a kind of a controlled 'framing error'. Very likely your USB/232
> product either cannot do a long-break, or is trying to simulate it in a
> manor the GE PLC doesn't accept.
>
> I cannot offer any advice on USB/232 that support 'long break'. Perhaps
you
> can set the PLC to use SNPX (or "breakless SNP). This might solve the
> problem.

SNPX normally requires a 3 character time break before sending an X-Attach message, so that probably won't solve the problem if that is indeed the case. I'd agree, the way to get around this is to use a breakless variant of SNP. They're normally used for modems, but I can see how'd they be useful here.

Alex Pavloff - apavl[email protected]
Eason Technology -- www.eason.com

A

#### Anonymous

> The problem is that the converter needs 5V that is supplied by the standard comm ports but not supplied by the USB serial converters. The 5V needs to be added from either an outside source or borrowed from another port that provides it. Hope this helps <

Sorry, but this is incorrect info. The DB9/RS232 side of the Horner mini-converter is not powered. The mini-converter sucks power from pin 5 of the DB15/RS422 connector on the PLC. This is how the old HHP (Hand Held Programmer) for the 90-20 & 90-30 was able to work without batteries. On the 90-70 CPUs, this is actually fused.

R

#### raywins

>.....One thing that may be causing a problem here is the BREAK character.... <

Thanks, I know that SNP over modems is a royal PITA, but I was hoping to be lucky. Someone out there has had to have done this by now!!?? If I get this to work, I will post solution.

R

#### raywins

>...I cannot offer any advice on USB/232 that support 'long break'. Perhaps you can set the PLC to use SNPX (or "breakless SNP). This might solve the problem. <

Lynn,
Prior reply suggested Digi may have an answer. I'm not sure if the programming port can be set for SNPX, but it's worth a shot. I'll check out your web site.
Thanks!

W

#### wang yunbo

The usb to RS232 adapter work well with windows-based software, which need not hardware interruption. But not for Dos-based software,
like LM90. So the solutions may be using VersaPro as program software or buying other adaptor which can set hardware interruption.

M

#### Mark Hensley (Kepware)

Hi Ray,

Just to be sure I gave the Digi International Edgeport USB serial product a try with our SNP driver and "Baddah Boom Baddah Bing" it worked.

I have the Edgeport 4 port version so you may need to look into other units if you need more or less ports but it does work with SNP.

I'm running XP Pro on my current PC but have also used the Edgeport with 2000.

R

#### raywins

That's the one thing I didn't try. Thanks for suggestion.

R

#### raywins

Thanks Mark,

Now to see what this will cost me ;-)
Digi needs to advertise the fact that this works with LM90. (Are you reading this Lynn?)

L

#### Lynn at Alist

> ...Digi needs to advertise the fact that this works with LM90. (Are you reading this Lynn?) <

Yes, I am - thanks to you & Mark Hensley for trying this. I am stuck writing new firmware for Digi Ethernet to Serial devices and don't get time to play with Digi USB (or even the 98% of Digi products I'm not supporting!).

There are just too @!$?$ many industrial protocols to be able to test them all in "meaningful" ways with every product - take as example the recent Modicon DTR/DSR issue where common data polling works fine with DSR low but
one cannot start some of the PLC if the PLC detects DSR as low? Who would have guessed such a distinction existed? I learned about it because our Modicon PLC here worked flawlessly with our product, but a large customer with some older, different model PLCs that couldn't start/stop. So testing with 1 or 2 of a vendor's PLC isn't enough even (I'm sure Mark/Kepware has a
similar "fond" opinion of testing) ;^]

This is one of the great values of the email forum.

Many Thanks
- Lynn