Alicat Mass Flow Control and RedLion

Hello, this is a long shot but worth a try. Has anyone interfaced an Alicat Mass Flow Controller with RedLion, using Modbus? I have placed several calls to Alicat and................ no comment. Their Modbus booklet shows an address of 42042 for Temperature, for example. Of course the Crimson software assigns the "4" for a hold register, but then requires a 6-digit address. I have tried different numbers for the holding registers, tried using gateway blocks in the Crimson software, I have tried putting in an offset, and I get nothing more than "---" for my values. I am able to point to Modbus registers in a Eurotherm Temp controller with the RedLion and all is well, so that confirms (in my mind) that the 2-wire RS485 from RedLion is fine. I don't know what else to try.

Try 402042 for 6 digit addressing.

Have you swapped the 485 data lines?

danw said:

Try 402042 for 6 digit addressing.

Have you swapped the 485 data lines?

Click to expand...

Thats another thing, the Alicat book that came with it shows 5 or 6 different DB9 pin outs. The one I have, isnt even listed in the book. I know I have the power wiring connected right.....at least its powered up. Lol. Yes Sir, I tried 402042...didnt work.

If it is wired right and talking and the device responds, you would get something besides "---" as a response.

It could be baud rate, wiring, termination, station address, missing termination resistors.

Once all those are right, you should get a value, but then you need to make sure that your byte and word order match up.

You don't have to use all 6 digits. Crimson is ok with 2042 in the address box. Check your device Modbus map and make sure it isn't using a zero address for the starting point. Example 40200. If so then add one to the address you put into Crimson. That has got me in the past.

OkiePC said:

If it is wired right and talking and the device responds, you would get something besides "---" as a response.

It could be baud rate, wiring, termination, station address, missing termination resistors.

Once all those are right, you should get a value, but then you need to make sure that your byte and word order match up.

Click to expand...

I set both baud rates the same. Something I dont understand is setting the device ID. I am accustomed to setting a number on the device. The Alicat gives me an "A-Z" or "@" for ID. I have tried changing that as well. I called our Alicat rep for support the help was very limited. I kept getting "it should be working" so I couldnt hang up the phone soon enough! So, I am not sure if its wired right, nor am I sure the drop number is set right. On top of that, the Alicat Modbus book shows 5-digit addresses whilst the RedLion (and Eurotherm), show 6-digits. I tried reversing order of bytes and words, with the Redlion. Thats why I was hoping someone here may have already crossed this bridge. With different pin outs for different Alicats, then having a model that matches none in the book that came with the darned thing.......

Bullzi said:

You don't have to use all 6 digits. Crimson is ok with 2042 in the address box. Check your device Modbus map and make sure it isn't using a zero address for the starting point. Example 40200. If so then add one to the address you put into Crimson. That has got me in the past.

Click to expand...

Crimson adds a digit, to make 6 digits. I tried different ways of addressing. Of course the leading "4" in Crimson is automatic for a holding register however, no matter what I do, another digit is placed, to make 6 digits. I have had other devices that had a certain offset to their address. For example, a Yokahama I set up many years ago I had to add "400001" to their modbus address, then have RedLion reverse the words, in order to use the Modbus addressing. That took a few days and a lot of frustration, to get running.

ceilingwalker said:

The Alicat gives me an "A-Z" or "@" for ID.

Click to expand...

I found a different description of slave ID. The Alicat Modbus manual here:
http://www.alicat.com/documents/manuals/ModbusRTU_Manual.pdf\
shows the slave ID node at register 40065. The bottom of page 3 shows the default ID is one.

Their hardware designation claims a 'send' and 'receive' line for RS-485, which is not true (it is true for RS-232).

2 wire RS-485 uses (+) and (-) lines. There is a Tx(+) and Rx(+) on the same terminal/conductor. Their understanding of RS-485 is goofy.

It isn't clear how you determine which options are installed. Since Modbus RTU and Profibus are options, as is the serial interface, I have to ask if you are sure that your particular model actually has Modbus RTU and are you sure that you know whether it is RS-232 or RS-485. If 485, their confusion leads me to re-iterate, have you swapped the (+) and (-) lines?

danw said:

I found a different description of slave ID. The Alicat Modbus manual here:
http://www.alicat.com/documents/manuals/ModbusRTU_Manual.pdf\
shows the slave ID node at register 40065. The bottom of page 3 shows the default ID is one.

Their hardware designation claims a 'send' and 'receive' line for RS-485, which is not true (it is true for RS-232).

2 wire RS-485 uses (+) and (-) lines. There is a Tx(+) and Rx(+) on the same terminal/conductor. Their understanding of RS-485 is goofy.

It isn't clear how you determine which options are installed. Since Modbus RTU and Profibus are options, as is the serial interface, I have to ask if you are sure that your particular model actually has Modbus RTU and are you sure that you know whether it is RS-232 or RS-485. If 485, their confusion leads me to re-iterate, have you swapped the (+) and (-) lines?

Click to expand...

Yes Sir, the "485" is in the part number, plus it has RS485 to set the coms settings, on the unit. This is why I am confused about the unit number being a letter (A-Z). I was able to set the baud rate to make it match the RedLion. I have that Modbus Operating Bulletin for the Alicat. I don't know how to determine which pins are really for the RS485. I tried a couple of configurations just to determine which ones were the 24VDC supply! I honestly don't think addressing is the fault, for now. I don't know how to tell which pins to use for the RS485 and as I recall, I believe the RedLion also needs for this unit to have a coms ground connected.

Make 12 devices in the Red Lion. Use drop numbers 1-12. Add one tag and copy it to each of them. Put all twelve tags on a screen. See what you get.

Swap the comms + and - wires. See if anything changes.

If the device has LEDs for the RS485 port, you may be able to tell something from them. When they are backwards, you may see only a receive LED or a send LED light up. When correct, you may see rapid blips on both of them.

Technically, 2-wire RS485 is supposed to be three wire if you counted the signal common.

1) Modbus is an option. RS-485 does not necessarily mean Modbus. It could mean 'streaming'. How do you know this specific unit talks Modbus?

2) Unless the default slave ID was changed somehow, it's still at one, the default.

If you have more than one device, then you'll have to worry about changing the slave ID, otherwise it should be one.

3) The flow controller has something called "streaming mode" (using RS-232/485) that uses the alpha ID letters to uniquely identify different devices. The manual specifically states that RS-485 units do not streaming mode (pg 33). So the alpha stuff does not apply.

4) Why do you think that the user manual that shows the pinouts for RS-485 is incorrect? Why aren't pins 3, 5, and 8 on the 8 pin DIN connector the RS-485 lines?


5) Polling slave ID 1 with Function Code 03 (read Holding registes) from the 2 registers starting at (4)04022 [or (4)04021 or (4)04023 when you test for a one offset addressing issue] should get a reply with the temperature value, probably in Deg C (unlikely to be a zero value).



If it were me, I'd be polling with Modscan so I could see the additional feedback, like "no connection" or "time out".

danw said:

starting at (4)04022 [or (4)04021 or (4)04023 when you test for a one offset addressing issue]

Click to expand...

grumble, grumble, can't copy correctly. Should be:

starting at (4)02042 [or (4)02041 or (4)02043 when you test for a one offset addressing issue]

danw said:

grumble, grumble, can't copy correctly. Should be:

starting at (4)02042 [or (4)02041 or (4)02043 when you test for a one offset addressing issue]

Click to expand...

Thank you Dan. I have all of this documentation. My Alicat is a DB9 and in the book that came with the unit, it shows 5 or 6 different configurations. That's my dilemma.

danw said:

).

If it were me, I'd be polling with Modscan so I could see the additional feedback, like "no connection" or "time out".

Click to expand...

Interesting. So, do you leave the unit connected to the RedLion when you run this s-ware?

ceilingwalker said:

do you leave the unit connected to the RedLion when you run this s-ware?

Click to expand...

I would not leave the Red Lion connected because establishing serial comm is always a *****.

There's lots of things that can trip up serial comm. If the Red Lion is connected, you have to figure out whether it is applying bias, whether it has termination built in, ya-da, ya-da.

I'd keep it as simple as possible, Modscan master on the PC with an RS-485 connection to the slave, the purpose being to prove that the slave communicates valid data from a given register.

The PC RS-485 connection can be either a USB/232 converter to a 232/485 converter or a USB/485 converter.