Shop OBEX P1 Docs P2 Docs Learn Events
towards a P2 PLC - Page 10 — Parallax Forums

towards a P2 PLC

14567810»

Comments

  • MicksterMickster Posts: 2,734
    edited 2025-01-28 05:11

    @evanh
    Hypothetical example:

    A company wants to introduce a product because they know that they can sell a minimum of 1000 units/year, right off the bat. The target selling price is, say, $US10K.
    The product involves 4-axes of servo-motion control and speed and precision are top priority.
    The controls engineer has decided that the Trio MC405 with servo option is the best bang-for-the-buck.
    I try to explain that the P2 would be far superior when it comes to cost/performance:

    MC405 Motor Command Resolution: 12bit
    P2 Motor Command Resolution: 16bit (or more with an external DAC.)

    MC405 Maximum Encoder quadrature counts/sec: 6M
    P2 Maximum Encoder quadrature counts/sec: 40M (conservative)

    MC405 Minimum servo-loop-time: 125usec
    P2 Minimum servo-loop-time: 16usec (conservative. Pretty sure I've hit 2.5usec)

    MC405 Support for 2 encoders/axis (aka: Dual-loop feedback): No
    P2 Support for 2 encoders/axis (aka: Dual-loop feedback): Yes

    MC405 shiny enclosure: Yes
    P2 shiny enclosure: No

    My Trio Price list is from 2018 and it shows GBP 1,280 for the MC405+Servo option
    P2 with buffering to suit: GBP 100?

    For his own comfort and job security (and ignorance), the engineer is gonna go with the shiny box option.
    He doesn't care that the company is losing > million bucks in profit to Trio.

    EtherCAT: No, @ManAtWork realised that it's not as "open" as claimed.

  • .> @Mickster said:

    @evanh
    Hypothetical example:
    ...
    MC405 shiny enclosure: Yes
    P2 shiny enclosure: No

    My Trio Price list is from 2018 and it shows GBP 1,280 for the MC405+Servo option
    P2 with buffering to suit: GBP 100?

    so... what you're sayin' is, it is time to make a shiny box. :)

  • MicksterMickster Posts: 2,734

    @refaQtor said:

    so... what you're sayin' is, it is time to make a shiny box. :)

    Absolutely...we need a Prop answer to the Arduino stuff.
    As @Rayman pointed out, 0-10v digital outputs? What use are those?

    Looking at the H7 Portenta Machine Controller; they have support for 24v encoders? Why not the ubiquitous 5v differentials?

    :s

  • JonnyMacJonnyMac Posts: 9,207

    0-10v digital outputs? What use are those?

    I don't know what they did with them, but my friends in the tech shop at Disneyland asked me to show them how to do 0-10v with a P1.

  • MicksterMickster Posts: 2,734

    @JonnyMac said:

    0-10v digital outputs? What use are those?

    I don't know what they did with them, but my friends in the tech shop at Disneyland asked me to show them how to do 0-10v with a P1.

    0-10v analog is a standard control signal but the standard for digital I/O is 24v

  • evanhevanh Posts: 16,145

    I'll repeat the question since you've avoided answering so far...

    @Mickster said:
    Some of the stuff that is merely parroted here is laughable.....straight from a Siemens/AB/Beckhoff brochure.

    Care to be specific instead of widely throwing abuse at nothing in particular?

  • MicksterMickster Posts: 2,734

    @evanh

    Bored with the thread. Out-of-date thinking.

    I'll be using jump-tables and tightly integrating motion with logic control :+1:

  • @Mickster said:
    @evanh

    Bored with the thread. Out-of-date thinking.

    Well, that is PLC.. it's an old, old, concept intended to easily convert relay logic to more modern control which would be understandable to anyone who could read a schematic... It just never had the dignity to die.

    We just won't have the raw processing power to include in the base unit, PLC like equivalent of things like:

    I'll be using jump-tables and tightly integrating motion with logic control :+1:

    Which is exceedingly cool, and surely the way to go if you learn the processor, but the whole PLC approach is to engage those who don't have the time to do so.
    It would be a very good thing to include code boxes to call native SPIN and/or/PASM to implement fancier stuff, and maybe adopt some of the better into a library..

  • evanhevanh Posts: 16,145

    @Mickster said:
    @evanh

    Bored with the thread. Out-of-date thinking.

    So you're just hurling abuse 'cos you can then.

  • MicksterMickster Posts: 2,734

    @evanh said:

    So you're just hurling abuse 'cos you can then.

    Didn't realise that those might be hurty words

  • evanhevanh Posts: 16,145

    Being insulting is an intentional act. There's no not intending to hurt.

  • RaymanRayman Posts: 14,876

    Found an RJ45 to VGA adapter on Amazon.
    Pretty sure it must be passive.

    This kind of thing makes that @refaQtor idea of just adding a few RJ45 connectors for things like RS232, RS485, and now VGA kind of interesting.
    What they actually do could be software controlled to some degree...

  • @Rayman said:
    Found an RJ45 to VGA adapter on Amazon.
    Pretty sure it must be passive.

    This kind of thing makes that @refaQtor idea of just adding a few RJ45 connectors for things like RS232, RS485, and now VGA kind of interesting.
    What they actually do could be software controlled to some degree...

    well... that would b e a trick.. and one that I'm not gonna attempt. the common things of all the pins/ports I had suggested used 2 or 4 pins andh had some pretty clear overlap.

  • TubularTubular Posts: 4,717

    Has any consensus been reached on form factor for a shiny case?

    I do like the bud enclosures but fear they end up looking a bit 'standard'. Its possible that making a really presentable top/membrane/display could really help lift things

    There's also a metal AL extrusion version of those bud enclosures, which would require CNC machining to expose terminals and display, but I think could come up really nice. I need to order one and see it IRL

  • TubularTubular Posts: 4,717
    edited 2025-01-29 00:13

    We could do worse than something like this for $10,
    https://www.aliexpress.com/item/1005007499342518.html

  • refaQtorrefaQtor Posts: 118
    edited 2025-01-29 01:36

    @Tubular said:
    We could do worse than something like this for $10,

    for me, just the bit that mounts on the DIN rail isn't the final package ... leaving exposed terminals until mounted in another giant box. I get that many/most will be used like that. I'm sticking with the "beaker" profile to fit in these boxes https://www.amazon.com/gp/product/B09TGDPRWH where I can have sealed fittings around cables in and out of the final enclosure with a clear sealed lid to expose some system status and easy access to buttons, etc. but, that's just me.

  • TubularTubular Posts: 4,717

    Ah yes, I remember the previous posting about those

    Fwiw Bud have videos on their site showing UV printing directly onto their enclosures, and also can cut/machine the openings. What I'm curious about is whether they can mill openings in the DMX AL extrusion so you can fit the standard 9 or 12 way 0.2" pitch terminal blocks like with the plastic series.

  • evanhevanh Posts: 16,145

    I prefer the card type layout. It's more compact from the get go, and is usually oriented for expandability. Downside is they are also usually more expensive. But that might be a marketing factor since they're also treated as more valuable.

  • refaQtorrefaQtor Posts: 118
    edited 2025-01-29 15:32

    @Tubular said:
    Ah yes, I remember the previous posting about those

    Fwiw Bud have videos on their site showing UV printing directly onto their enclosures, and also can cut/machine the openings. What I'm curious about is whether they can mill openings in the DMX AL extrusion so you can fit the standard 9 or 12 way 0.2" pitch terminal blocks like with the plastic series.

    I suspect they offer such a service. and when I get to the point of wanting 100+ identical, I might investigate. now, I'm happy enough with the flexibility of the open edges of the "breaker" profile.
    The metal ones are nifty. but, I don't care to have that breaker profile unit so heavy duty, as it, in my case, will always live in an external close fitting enclosure.

    If I was going for generic PLC, I guess I'd go with the larger edge mounted gang-able block format of the big guys. That is maximum flexibility for an ample large box with many other components.

    The flat mounted DIN rail board enclosures (like you just noted) seem to me to be a good fit for a single purpose component like a dedicated motion or thermal or custom equipment control that may, or may not be enclosed in another box.

    On my external breaker panel enclosures, each one is different with different "drill-outs" (in the plastic "knock-out" locations) with RJ45 and USB-B pass bulkhead pass-through connectors and cheap short RJ45 patch cables. I have a punch for DB9 connectors. still experimenting with the options.

  • RaymanRayman Posts: 14,876

    How do you all terminate the wiring?

    Seem to recall the wires we use having some kind of solid ends so that you stick in the terminals.
    Tried to look for something like that, but couldn't find it...

    Be nice if you could buy wires with that already on, but not seeing that either...

    Might be getting a little ahead of myself though, still need to make the thing...

  • evanhevanh Posts: 16,145
    edited 2025-01-29 23:07

    Bootlace ferrules for lighter wiring in the clamping terminals. Crimp lugs for the heavy stuff that gets bolted down.

    EDIT: Spelling correction! Thanks Jon.

  • RaymanRayman Posts: 14,876

    Thanks @evanh "bootlace ferals" is exactly what was thinking about. Would never have guess that name...

    Looks like going to need to buy a new tool for those...

  • evanhevanh Posts: 16,145
    edited 2025-01-29 23:15

    @Rayman said:
    Looks like going to need to buy a new tool for those...

    Don't have to, they're soft. When I'm just quickly patching things I use my side-cutters to pre-crimp the ferrules. But you do want to use a good screwdriver that can do the terminal up tight without burring the screw head.

    Most importantly, the ferrules have to be notably smaller than the terminal. Otherwise you can end up with the ferrule being clamped tight but the wire still loose inside the ferrule.

  • JonnyMacJonnyMac Posts: 9,207

    Looks like going to need to buy a new tool for those...

    When I was doing some work with my friends at Disneyland they showed me their $1200 tool for ferrules -- a bit out of my budget for an infrequently used tool. But then, I came across this and haven't had any problems. When I use #18 and #20 stranded for big displays it's a lifesaver -- no more tinning wires or having shorts caused by frays.
    -- https://www.amazon.com/gp/product/B073TZ5BBG

  • evanhevanh Posts: 16,145

    Nice! That's a great buy Jon.

  • @Rayman said:
    Thanks @evanh "bootlace ferals" is exactly what was thinking about. Would never have guess that name...

    Looks like going to need to buy a new tool for those...

    this sort of thing.
    tool is cheap and does a nice job effortlessly.
    I'd found that, the smallest ferrule narrow enough to fit in, is too short to make good contact in spring terminal blocks.

  • evanhevanh Posts: 16,145

    Oh, yeah, spring terminals, they are quicker, for removal at least, but I'm not a huge fan of them. They work best without ferrules.

  • One trick useful here is to buy the twin version of the bootlace ferrules, because they have an oval shape and can nest side by side when going into terminal blocks, both screwed and spring terminal varieties. See the 'diameter' column in the attached and you'll see what I mean.

    While these are really designed for two wires you can just go down a size and use them with a single wire

Sign In or Register to comment.