Novamill communication issue

[jgoodman]

I am taking robotic integration this semester and am part of the team that is taking care of the cnc milling portion of the project. Right at the beginning of the semester our pc died and is not recoverable (looks like motherboard caps went out and it hurt more than just the motherboard). So now we are having to use one of the laptops in our lab as we do not have a spare pc. I will put several pictures up at the end to show everything we are currently using. Also this is my first time ever using the Denford Novamill and the VR Milling v5 software.

So as of right now we have purchased a new copy of the VR Milling v5 and have downloaded it on the laptop. We also ordered an adapter cable to go from usb to rs232 (this one to be exact: https://www.amazon.com/Adapter-DNCLINK- ... pc&sr=1-13). I have tried several different things on the laptop side; ftdi drivers, COM port settings, played with settings in the milling software. Still getting the usb/rs232 connection error when trying to establish connection to machine.

A change in pc is out of the question right now but change in cables is doable. Questions are welcome but I wont be able to check on much physically until Monday.

Any help is much appreciated,
jgoodman

[Martin]

What machine have you selected in VR Milling?

[TDIPower]

A real quick one you could try is swap pin 2 and 3 over on the chassis mount rs232 on the cab.

I have over the years met many counts of RS232 issues which come back to an incorrect wiring. Now one would have thought that as the machine was working before and you are using the same lead it should be fine BUT it all depends upon how the USB to RS232 has been wired.

It would be worth trawling through the RS232 issues on here, looking at the chassis mount connector it looks like it has more that the TX/RX ground so will need the corresponding connections in the RS232 adapter or loop backs in the lead. You may find the USB adapter doesn't even connect to more than the TX/RX/ground as so much RS232 dropped all the RTS CTS and other connections.

There is a 'god lead' pin out somewhere which is 9pin to 9pin and goes direct to the board but you shouldn't need that as you had it all working with the lead you have.

So my bets are either pin 2/3 swap or you need an rs232/usb that uses all the connections.

[DavidB]

One thing that can cause comms issues with some software, I don't know if it effects Denford software, is the latency setting in the FTDI driver, it's likely in the advanced settings dialog on the driver page you posted earlier. I think in order to use the USB efficiently it waits in case more characters come along before sending them on the USB bus but with software I used to maintain it could lead to timeouts so the latency had to be set to the minimum in order to prevent that.

[TDIPower]

There has been some posts in the past about usb/rs232 and issues over unreliable connection, i dont recall complete failure. I could well be wrong. I seem to recall some types work some dont and it being down to the chip used in the converter. Im going back about 5yrs on this now, i have used 2 with success in the past.

[jgoodman]

What machine have you selected in VR Milling?

I have been going to the legacy machines and selecting Novamill as my machine.

I wont have access to the machine until 8am Monday but I definately want to give your ideas a try @TDIPower.

@DavidB like you said I have been messing with the latency settings a bit but I had no idea as to what I should set everything to. I don't remember if we tried going super low so I will also give this a try Monday before tampering with the pins.

Thanks for the suggestions hopefully I can get some progress made next week.

[Martin]

If you look in the VRMilling directory on the computer you can try Mintfinder to see if that will connect.

[jgoodman]

Between yesterday and today I have tried most of the ideas posted here with little to no success. I tried changing the latency in the COM advanced settings and never came up with a good result. I switched pin 2/3 as TDIPower suggested and that was also a bust. When I booted up MintMan and tried to establish a connection it said "Comms not OK".

The only thing I haven't tried is a different cable so are there any usb/rs232 adapters approved by Denford that they know work or @TDIPower would it be able for me to get a usb/9pin serial and plug directly into the controller board.

[DavidB]

Is there a battery below the LED, it looks like there might be but the image doesn't have enough detail to be sure. Worth checking for any appearance of deposits on the PCB around the battery as they often leak electrolyte after a time and it typically effects the comms signals. Can you find one of these in the UK https://www.ebay.co.uk/itm/DB25-Male-to ... 4346634474 . Do you have access to an oscilloscope? Was the Novamill attached to the PC when it died?

Might be worthwhile having a look at serial comms monitoring programs some mentioned here https://www.virtual-serial-port.org/art ... utilities/ . One USB RS232 converter I have had one with the installation and it was handy for debugging a particular nasty bug. I don't know where they hook the system so it would be handy to have a verification of signal activity on the lead.

This might be handy https://www.ebay.co.uk/itm/Serial-Loopb ... 3197010934 or you could make your own according to its details. It should enable you to see if data is being sent out and returned by using Hyperterm or similar terminal software to show the characters typed as they're sent out and back. If you try that links paper clip trick to bridge TX and RX make sure you're grounded to the chassis in case you're carrying a static charge as I knew someone that didn't and took out not just one serial port but the other and a number of other PC peripherals.

[TDIPower]

I just dug out my 2 USB to RS232 devices and plugged them in to my desktop PC running win 10,
The 1st is a Prolific PL2303HXA which interestingly Win to reports in the comports as

PL2303HXA phased out since 2012 Please Contact your supplier

Now I was running this in windows XP when I used it in DOS box as the machine was a DOS based machine.

The other is reporting the same (I'm typing whilst testing)

These USB to RS232 adapters were both used on a Win XP laptop that did not have RS232 until I got a PC with 2 RS232 ports.

I use the onboard RS232 ports on a proper PC rather than USB to RS232 as they are still on motherboards you just need the breakout, but understand you only have access to a laptop.

It's that matter of you need to prove where the fault is. Getting hold of a PC with onboard RS232 would be the best way just to work out where the issue is.

Just a thought. I know I did not change from 9600 bits, 8 data bits, Parity none, Stop bits 1, flow control none, In advanced use FIFO 16550 was selected and the TX/RX buffers set to high 16/14. When I did this and those are the settings my USB adapters default to on plug in.

The direct 9pin to 9pin connections are (nc = not connected)
1 DCD nc 1 DCD
2 RD - 3 TD
3 TD - 2 RD
4 DTR - 6 DSR
5 GND - 5 GND
6 DSR - 4 DTR
7 RTS - 8 CTS
8 CTS - 7 RTS
9 RI nc 9 RI

[TDIPower]

Those RS232 LED units are great, I got my Mum to send me mine over (I moved out years ago but knew it was still there) when I was sorting things out with another machine, I had one of the ones you use jumper cables to link the pins but that had vanished. Makes it a lot easier to ID and try things.

Pete