Density vs Termination

[BobO]

We tend to evaluate these things strictly from a system positioning perspective, i.e. we need a system to fit in a particular market space, even if it creates compromises. Help us understand what your preference would be for the expansion modules (up to eight) to be added to a DL06 class brick. And remember...this is hypothetical...

[Mike Nash]

Tough choice! (I had to download the Autocad file on one of the D0-06 modules just to see which it was.) The 3.8mm is usable, but not so much with 18AWG MTW. So I guess I would trade I/O points for usable 5mm terminals. But I like cage clamp screw terminals. No one-shot chintzy stuff please.

[BobO]

DL06 is probably 3.5mm. Same screw size as 3.8mm, but even tighter pitch. They are tiny, either way.

[Controls Guy]

Personally I think MTW/TEW is overkill for I/O anyway.   I typically use UL1007 (1/64" insulation, PVC, 300V) or 150V multi-conductor cable.   In a low-voltage (120 or 240V) panel, it's obviously OK,  and in a 480 panel, power and I/O are generally well-enough segregated you don't need to rate for 480.  Think about the cables they use for ZipLinks, for example.  Definitely not 600V.

[jwbaker3]

For me on most project panel size (and cost) is an issue, so I prefer the higher I/O count. As long as there is a Zip link for the card. I mount the zip link with the terminal blocks for field wiring so they save enough labor on a panel build to offset the cost. If there is not a zip link I would want the lower I/O count with larger terminals. Without the zip link the labor to wire the small terminals offsets the cost of the larger panel.

JW

[BobO]

For me on most project panel size (and cost) is an issue, so I prefer the higher I/O count. As long as there is a Zip link for the card. I mount the zip link with the terminal blocks for field wiring so they save enough labor on a panel build to offset the cost. If there is not a zip link I would want the lower I/O count with larger terminals. Without the zip link the labor to wire the small terminals offsets the cost of the larger panel.

Doesn't larger size required by zip links offset the increased I/O density...at least if it fits in the local base?

[Controls Guy]

Agreed.  Getting higher point count per slot is pointless if a ZipLink is required because the ZipLink takes more room than another module.  People say ZipLinks replace the terminals you'd otherwise use and you have to factor that for cost and space, but if you're willing to take your field wires to a prefab device vs. a terminal per se, why not just wire to the I/O module?

[Mike Nash]

Personally I think MTW/TEW is overkill for I/O anyway.   I typically use UL1007 (1/64" insulation, PVC, 300V) or 150V multi-conductor cable.   In a low-voltage (120 or 240V) panel, it's obviously OK,  and in a 480 panel, power and I/O are generally well-enough segregated you don't need to rate for 480.  Think about the cables they use for ZipLinks, for example.  Definitely not 600V.

While I completely agree, many times where I am applying a small PLC is a retrofit and I am unusually fortunate if the existing wiring is 16AWG rather than 14AWG. 

[BobO]

Personally I think MTW/TEW is overkill for I/O anyway.   I typically use UL1007 (1/64" insulation, PVC, 300V) or 150V multi-conductor cable.   In a low-voltage (120 or 240V) panel, it's obviously OK,  and in a 480 panel, power and I/O are generally well-enough segregated you don't need to rate for 480.  Think about the cables they use for ZipLinks, for example.  Definitely not 600V.

While I completely agree, many times where I am applying a small PLC is a retrofit and I am unusually fortunate if the existing wiring is 16AWG rather than 14AWG. 

Ouch. The 3.8mm terminals we are looking at will support a 14AWG wire, but it isn't ideal.

I'm seriously considering 10 point and 16 point. It'll cost a bit money to tool the 10 point shell, but I'd really like a solution for those that don't require the extra I/O and who would prefer 5mm where possible.

[Controls Guy]

While I completely agree, many times where I am applying a small PLC is a retrofit and I am unusually fortunate if the existing wiring is 16AWG rather than 14AWG.

I did some mods not too long ago on an old machine with a 305 with IIRC 16-point modules.  The original machine builders wired the I/O in 14AWG THHN!  The individual terminals were perfectly happy with #14, but a bundle of 16-18 #14's didn't really fit the wire cavity on the module front very well.  [facepalm]

[Controls Guy]

Even the barrier-style terminals on the built-in points on an 05/06 are ridiculously tiny.

Issues like these are why I like the SLC/405ish form factor.  Hope we get there eventually!

[BobO]

One of the the things that has driven this conversation has been the observation that 20 x 14AWG is a jumbo sized bundle. I proposed that we limit the count for both reasons: larger terminations and pain of managing big wads of hatefully large wire. I lost.

The more I ponder, though, the more I like having both. To limit cost, we have molded some LED management features into the module housing, which would prevent using the same housing for both 5mm and 3.8mm terminations...at least if you want the status LED next to the associated screwhead. That said, having a 5mm housing and a 3.8mm housing might well be worth the cost of tooling. I think we could actually fit 12 points at 5mm. That would still allow for 132 I/O on the local base, and I find it hard to believe that won't do the bulk of our apps.

As for a 405-sized platform, it might happen. If the small footprint platform works well and there is a reasonable expectation of return on a high end platform, I would definitely like to do that next.

[Controls Guy]

At most, I'd give the LED's-next-to-the-screw issue a "mildly desirable".  I think the traditional 205/405/SLC way of having them all in an array at the top of the module is perfectly workable, especially if it helps you with some other compromise.

What I'm NOT crazy about is not enough LED's, leading to the A/B switch on the 16-pt 205 modules.  As long as there are LED's for every point, and they're laid out in a logical array pattern, it's not absolutely necessary for them  even to be labeled for point no.

[ATU]

Have you seen mPPE?  http://www.digikey.com/classic/documentredirector.aspx?doc=http://www.digikey.com/Web%20Export/Supplier%20Content/AlphaWire_216/PDF/AlphaWire_WP_EcoWire.pdf
It costs about twice as much as standard MTW, but has a higher temp rating, higher dielectric strength, more flexible and the insulation thickness is considerably less. One of my customers showed me it today. He uses it to wire high density I/O modules only and to him its worth the cost and is better for the environment.

[Controls Guy]

Off topic:  I'm in the middle of doing a significant expansion of a SLC installation.  I suggested starting over with Do-More or current AB offering, but he wanted to stay with the SLC.  (4) 13-slot racks.  I'd forgotten that classic AB RIO was designed by the Marquis de Sade.  Half racks, 1/2 slot vs. 1 slot vs. 2-slot mode, discrete mapped vs. BT, logical racks, G-files, max 32 groups, have to alternate 16-point modules in/out/in/out in 2-slot cause there's only one word in and out per group, etc, etc, etc, ad nauseum.  What a pain.  TI 5x0/5x5 stuff was always so much easier.  Run the wire, dial in the rack ID, power up, and you're done.

Part of it is that the SLC RIO scanner is so anemic compared to the PLC5, but still, the overall design is just so unnecessarily complicated.