Serial Cable question....

[maginter]

Has anyone run their bot wit a 25' Serial cable from the computer to the control box? I am remoting my control box and am currently running a 10' cable successfully. However when I go to a 25' cable the system starts showing "Data Error" when it tries to connect. I am guessing that I might have a bit of ambient electrical noise. The next step is for me to sheild the cable, but I thought that I would ask if this is a common problem. I am running a Botrods controller.

TIA,

MArk

[knight_toolworks]

12' is the max for a usb cable. it's best to get the computer closer. but if you must you want a powered hub in the middle of the run to kelp out.

[Brady Watson]

15' is the limit on USB without a 'repeater' to bump the signal strength up. 12' is the limit on a ShopBot in my experience. 15' or more will fail.

-B

[gabepari]

I seem to be an anomaly on certain things


I ran 27' (three 9' cables) on USB with success for about 9 months on my PRT-Alpha. Bone stock $300 Dell, on-board USB port, but expensive cables. It worked ok on cut files. Had one small hiccup that would show up on large pocketing routines (spoiler board surfacing). When I installed the 3:1 gearboxes, LOTS of problems... After rearranging the shop, I now have only one 9' cable, works much better.

Try to run a single 10' or less cable, and spring for a nice cable, makes a difference.

[Brady Watson]

The reason you had trouble with the 3:1 gearboxes is because you now need 3X the pulse stream to go the same speed....

-B

[maginter]

Guys -

As stated in the original post this is a Serial cable (9 pin) not a USB. The manual states that I can run as long as 50 feet. No joke, page 45.

Are you guys suggesting that I move to a USB to Serial adapter?

I want to run e chain for the bot.... older bot, and have the controler mounted on the table in a NEMA box.

[garyb]

Mark said serial cable not usb
I have a 25' serial cable on my prt and it runs all day long with no problems.
Gabe your not alone, I also have usb remote controllers connected to both of my machines which use usb, one has 21' of cable and the other has 31' of usb and they work fine so there are exceptions.

[maginter]

Hey Gary -

Do you know the manufacturer of your serial cable?

[khaos]

8849 I have used these guys exclusively for the last 10 years or so. I have found the prices to be great and the staff to be quite knowledgeable. http://www.norfolkwire.com

JMHO,

[maginter]

Thanks.....

I bought a Vaster DB9 25 ft. Actually, I bought two. But you know what. The more I look at them the more I think that they are knock offs. They look like Vaster ends, but the amazing part is that Vaster doesn't have a 25 ft M/F DB9... Go figure

[garyb]

Not off hand Mark but will check tomorrow for you

[richards]

Serial cabling is a black art. IF the cable uses a drain wire that is connected to the shell of the D-connector ON ONE END ONLY, IF the power supply actually gives +/- 25V (instead of the more usual +/- 5 to 12V), IF the baud rate is moderate (9600 baud or less), THEN you might be safe up to 50 feet. However, in an electrically noisy industrial workplace with lots of generated noise from the router/spindle as well as the pulses going to each of the stepper motors, you're going to have an excellent chance of picking up a lot of electrical interference that will superimpose itself on the serial pulse stream.

In an office environment, I've successfully run 9600 baud serial lines more than 100 feet. In electrically noisy shops, I've had trouble at 10 feet. At higher baud rates, the problems multiply.

[harold_weber]

Mike could you comment on whether or not all serial cables are shielded and whether shielding would help?

[garyb]

My mistake Mark, I'm running on lpt 25 pin not db9

[richards]

Harold,
Not all cables are shielded. In fact, very few of the cables that I've purchased were shielded. What I normally did was to buy the shielded cable and then make my own cables - not just to get the shielding, but also because there were too many 'standards', none of which matched my exact needs.

The RS-232 standard called for a signal range of +25V to -25V, although it tolerated a signal as low as +3V to -3V. The higher voltage range helped to minimize the cross-talk caused by electrical interference.

Unlike more modern protocols, like EIA-485, which use twisted pair cabling, the RS-232 did not require twisted pair cabling. On older devices that were limited to about 1,200 baud, the RS-232 performed very well. But, as speeds increased, so did the problems. Today, RS-232 has been replaced by USB/Serial or TCP/IP networking. Both of the newer protocols allow much faster data transfer with fewer problems.

If you decide to build your own cable, verify that you only have the shield connected on one end of the cable. If you connect the shield at both ends of the cable, you may get a 'ground loop' which can be fatal to your equipment.