Wiring, general electricals and earthing

Discussion in 'Motion System' started by yngndrw, Jan 1, 2020.

  1. yngndrw

    yngndrw Well-Known Member

    Joined:
    Dec 26, 2019
    Messages:
    55
    Likes Received:
    19
    Background

    I am not a qualified electrician nor do I work in the electrical profession - I'm a software engineer. My dad is however an electrical engineer and when I were younger I used to help him build electrical control panels, some of which controlled 30kW three-phase motors. This included both reviewing the drawings and wiring the actual panels. (Except for the high current cables, I simply wasn't strong enough to loop a 30mm^2 wire through a current transformer!) Needless to say my advice here is offered "as is" and without warranty or liability. I've given you information about my qualifications and experience so that you can make an informed decision on whether or not to follow my advice.


    General Wiring

    High current connections should be crimped and not soldered, especially when there is vibration or movement in the wires.

    Pliers are not crimpers and the flat tools with a simple crushing end are not crimpers. The only type of crimpers you should be using for the red, yellow and blue connectors are ratchet crimpers. After crimping a connector on a wire, give it a good pull - If it can't take a decent tug, you should re-do it.

    Red crimps are for 0.5mm^2 to 1.5mm^2 wire, blue crimps are for 1.5mm^2 to 2.5mm^2 wire and yellow crimps are for 4mm^2 to 6mm^2 wire. Always use the correct coloured crimp for your wire size.

    Bootlace ferrule crimps should be used for screw terminals rather than bare or tinned wire - Again use the correct ratchet crimper and not pilers. These are also useful where you need to connect two wires into the power supply, where two ring crimps may not fit.

    Strain relief should be provided at both ends of a moving wire - The connections should never be subject to movement. I personally like to use silicone wire with heatshrink around the areas where I'll be fixing it. I then use cable ties to attach these areas to the frame.

    Don't run high voltage wires along side low voltage wires - Ideally the heatbed should have separate drag chains for the power wires and the thermistor wires.

    Don't run high current wires along side low current wires - Ideally the thermistor wires and heater wires would be separated for the hotends, although another option is to use shielded cable for the thermistor. This is even more important when using a thermocouple, where the voltages are minute.


    SSR

    The Omron G3NA-210B-UTU SSR is to be derated by 30% when not mounted on a heatsink which gives a good margin when used in a 240V region but is right on the limit when used in a 110V region. Additionally, this rating is given when the SSR is mounted on a thermally conductive surface like the side of a control panel. As such, mounting the SSR directly onto the plastic panel may be a fire hazard in 110V regions: "If a material with high thermal resistance, such as wood, is used, heat generated by the G3NA may occasionally cause fire or burning. When installing the G3NA directly into a control panel so that the panel can be used as a heat sink, use a panel material with low thermal resistance, such as aluminum or steel."

    The heatsink tab on the back of the SSR is not earthed and metal screws are used - The screws should therefore be earthed.


    Earthing

    Your earth wiring should be continuous where possible, an example of this is where the guide suggests that the earth feed from the IEC connector is taken into the earth terminal of the power supply, but the earth to the rest of the machine is then taken from the mounting hole on the side of the power supply - These should both connect into the earth terminal. Another example is the heatbed earth where the guide shows them both connecting to different earth screws - They should use the same earthing point and not rely on the conductivity of the heatbed bracket.

    If you have many earths to connect up, it is often better to have a single earth stud and use a series of ring crimps to connect everything to this point - A good place to put an earth stud would be on the aluminium frame.

    Spring washers should be used on the earth screws - Firstly to prevent them from coming loose and secondly to ensure that there is good connectivity with the underlying metal. This is especially important when earthing the aluminium frame and the coated heatbed. The coatings should be scratched off even when using a spring washer.

    Always confirm earth continuity with a multimeter for any exposed metal service, including bolts.


    Unrelated To Electricals

    Although this is common practice in the 3D printing world, the Z-axis lead screw looks to be axially supported by only the motor's radial bearings - As such these bearings must take the entire weight of the heatbed, the part being printed and any forces due to acceleration / deceleration. This is not good practice and also results in an undefined amount of z-axis slop. Lead screws and ball screws are meant to be both axially and radially supported at the driven end and radially supported at the floating end. (Using preloaded back-to-back angular contact bearings on the driven end.) Just because it's common practice (Probably due to designers often reusing sections from other machines) doesn't mean it's a good idea.
     
    #1 yngndrw, Jan 1, 2020
    Last edited: Jan 2, 2020
    Aluminati, Nibbels and mebrein like this.
  2. blarbles

    blarbles Well-Known Member

    Joined:
    Aug 10, 2019
    Messages:
    108
    Likes Received:
    68
    Is using the SSR to get 60C - 75C on your heat bed going to put US users without heatsinks at any real risk or are we talking maxing out your heatbed over a 20hr print?
     
  3. yngndrw

    yngndrw Well-Known Member

    Joined:
    Dec 26, 2019
    Messages:
    55
    Likes Received:
    19
    That's a really tricky one as there are so many variables. For reference, here's the datasheet for the SSR: https://www.mouser.co.uk/datasheet/2/307/g3na_ds_e_12_1_csm165-1221238.pdf (I took the part number from the photo in the store, it may be worth verifying that this is the same part as the one you're using: G3NA-210B-UTU)

    The factors to consider are:
    • The actual voltage from your supply - In practice this tends to be slightly higher which for a fixed resistance heater results in a slightly higher current
    • The duty cycle of the heater, I.e. On time vs off time - This itself depends on the ambient temperature, the target temperature, thermal losses of the heat bed, whether or not the underside of the heat bed is insulated
    • The safety factor that the SSR manufacturer has built into their specification
    The supply voltage really only affects the initial heat up as when up to temperature, the higher heater power will reduce the duty cycle. I used 110V for my calculations but if you know your supply voltage you can substitute it by first calculating the nominal resistance (10 Ohms for the 110V / 800W version) and then calculating the current from that. At 120V it would be 8A.

    Duty cycle is the easiest to start off with as it's easily measured if you have the printer setup. The duty cycle will increase as the target temperature increases and as the ambient temperature decreases, so it's best to work with the worst case scenario in which you'll be using it. The duty cycle can be used as a multiplier against the load, so if it's on 70% of the time (A number I'm plucking from thin air, you'd have to measure it) then the load when up to temperature will be 0.7 * 7.3A. It's worth noting that the load when initially heating up is constant so that will be the full 7.3A.

    Omron is a reputable and highly regarded manufacturer so they are likely to use a decent safety factor when rating their products. Having said that, looking over their datasheet again their load graph and table gives a different rating to the 30% derating that they mention in another section - It turns out that the 30% derating is when using the heatsink with the airflow partially blocked. I've attached a picture below and the graph shows this same 4A limit for ambient temperatures between -30C and 40C:
    Omron G3NA-210B-UTU Load Current.png

    Unless I've missed something or made a mistake, this SSR is being run at double its rating for 110V regions. I'd suspect the duty cycle to be quite low once it's up to temperature, but I wouldn't want to be running it at this level for the warm-up period and if I were in E3D's shoes I certainly wouldn't want to recommend this setup.

    The matching heatsinks are Omron Y92B-N50 or Omron Y92B-A100. Please remember to earth the heatsink if you add one.
     
  4. Joe Pomo

    Joe Pomo Well-Known Member

    Joined:
    Sep 21, 2019
    Messages:
    109
    Likes Received:
    41

    Thanks for the heads up about the SSR (I'm using 110V power). What do you think about mounting the SSR directly to the machine's bottom aluminum plate? Also, I haven't read the docs, but is thermal paste recommended between SSR and heatsink?

    Thanks again!
     
  5. yngndrw

    yngndrw Well-Known Member

    Joined:
    Dec 26, 2019
    Messages:
    55
    Likes Received:
    19
    I was wondering how best to fit the heatsink in and hadn't thought of that, but you're absolutely right - The datasheet explicitly mentions an "aluminum plate measuring 150mm x 150mm x 3.2mm" as an alternative to the heatsink and the bottom plate far exceeds that so it's ideal. I'd explicitly earth the bottom plate rather than relying on the continuity between the aluminium extrusion and the bottom plate though.

    You probably don't need thermal paste, but I'd advise it. Please keep in mind that the job of thermal paste is to fill the imperfections in the surface of the metal as the air bubbles would form a thermal insulator. Metal on metal contact is a far better thermal conductor so you should use the bare minimum amount of thermal paste, put some on the SSR and then scrape it flat with a card.

    Edit: Just noticed that the high temperature heat bed documentation does mention the heatsink, but sadly this seems to have been missed out from the tool changer documentation and the tool changer store page.
     
    #5 yngndrw, Jan 3, 2020
    Last edited: Jan 3, 2020
  6. Joe Pomo

    Joe Pomo Well-Known Member

    Joined:
    Sep 21, 2019
    Messages:
    109
    Likes Received:
    41

    Thanks for the feedback! I used the existing air vent slots in the plate. The holes didn't line up perfectly with adjacent slots, so I printed a small clamp in nylon for the other hole. Feels pretty solid.
     

    Attached Files:

    Nibbels likes this.
  7. yngndrw

    yngndrw Well-Known Member

    Joined:
    Dec 26, 2019
    Messages:
    55
    Likes Received:
    19
    Looks great, almost like it was meant to be mounted like that - I think I'll mount mine in the same way.
     
  8. Aluminati

    Aluminati Member

    Joined:
    Dec 27, 2019
    Messages:
    7
    Likes Received:
    3
    I think this is a valuable contribution and concern that needs to be addressed when building the Tool Changer. I am no electrician, and reading this post strains what little understanding I have on electrical matters, and I am sure I will prove that with this post. But, according to the chart load rating is from 24v to 240v AC, load rating range is from 19 to 264. Then we have the allowable amperage with and without the heatsink. With the heatsink , up to 10 amps output. Without the heatsink, up to 4 amps output. The way I am reading this, the SSR is only rated to 4 amps output without the heatsink regardless of whether the incoming voltage is 24v, 120v, or 240v.

    On a more practical note, has anyone just put an IR thermometer on it to see how hot it gets? To me, that would also be a good indication of whether a heatsink is needed. On that note, what is the actual concern? That the internals of the the SSR will get too hot and cause a fire or that the metal back plate will get too hot and melt something or cause a fire?

    Finally, don't you love the wording? "If a material with high thermal resistance, such as wood, is used, heat generated by the G3NA may occasionally cause fire or burning." Look guys, if it only happens every once in a while, can it really be that bad?
     
  9. yngndrw

    yngndrw Well-Known Member

    Joined:
    Dec 26, 2019
    Messages:
    55
    Likes Received:
    19
    Yes that's correct, the rating is based on just the current and not the voltage. The current is set by the resistance of the heat bed which differers based on the supply voltage in order to give an 800W output in both regions.

    The datasheet states "Heat generated from an G3NA error may occasionally cause the output element to short, or cause fire damage." - This essentially means that it is possible for the SSR to fail in a closed or "on" state, leaving the head bed at full power until the heatbed's thermal cut-out prevents it from reaching a silly temperature. (I hope they have a built-in thermal cut-out, the documentation provided doesn't say either way - I wouldn't want to rely on it either way)

    Aside from that, being mounted on Acrylic (I think) with a melting point of 160C, it could possibly melt that and fall off. It could start melting wires.


    It's all what-ifs, coulds and maybes - But the long and short of it is that there's a fire risk.

    Anything to cover their back while trying to sound safe. :)
     
  10. Nibbels

    Nibbels Well-Known Member

    Joined:
    Dec 12, 2019
    Messages:
    165
    Likes Received:
    32
    Screenshot_2.jpg Screenshot_4.jpg Screenshot_5.jpg Screenshot_6.jpg
    I clamped some old metal part behind it. When I drive my bed after switching the PSU I can easily tell if there is heat. I will connect a mass to it later.
     
  11. yngndrw

    yngndrw Well-Known Member

    Joined:
    Dec 26, 2019
    Messages:
    55
    Likes Received:
    19
    What's the pad that's stuck between the metal and the SSR? Is it some kind of thermal tape?

    Good solution if you want to keep the original location, just keep in mind that the datasheet calls for a 150x150mm plate for the full rating and that's probably assuming air movement on both sides.
     
  12. Joe Pomo

    Joe Pomo Well-Known Member

    Joined:
    Sep 21, 2019
    Messages:
    109
    Likes Received:
    41
    I've seen others use thermal cutoff fuses like this:

    https://www.digikey.com/products/en...146&quantity=&ColumnSort=0&page=1&pageSize=25

    Something similar is probably a good idea, but may require a custom metal fixture + thermal potting compound.
     
  13. Nibbels

    Nibbels Well-Known Member

    Joined:
    Dec 12, 2019
    Messages:
    165
    Likes Received:
    32
    Thats some thermal tape which I had from a bunch of PC RAM. All this should do is to deliver the heat to the aluminium part. If I then recognize the metal gets a temperature > 50..60°C I will do more. But I dont think so: Because the part stays alive with the whole thermal-connection-side covered by plastic (original state in the TC)
     
  14. yngndrw

    yngndrw Well-Known Member

    Joined:
    Dec 26, 2019
    Messages:
    55
    Likes Received:
    19
    Certainly something to consider, although it would be a bit of a pain to do and you'd probably want a few of them in strategic locations. It would be good to find out if the heatbed already has protection for excessive temperatures before starting on such an endeavour, does anybody know if it has a cut-off or not?

    Ah I see, I've not seen one so thick before.

    It's very possible that Omron have designed an excessive safety margin into their ratings, but I wouldn't trust running a part at over double its rating. If I were in E3D's shoes, I certainly wouldn't be comfortable recommending this setup without extra cooling as they would be liable for any fire damage or injury caused. Sadly the SSR datasheet is quite limited so we can't find out much more, although they do specify a 90C limit for the optional temperature cut-off on their larger heatsinks so that should be used as the absolute maximum temperature. Will be interesting to see what temperature it actually reaches - Especially with a high bet temperature in a drafty environment.
     
  15. yngndrw

    yngndrw Well-Known Member

    Joined:
    Dec 26, 2019
    Messages:
    55
    Likes Received:
    19
    I've just confirmed with E3D's support team, the high temperature heated beds do not include a thermal cut-off.
     
    Nibbels and Joe Pomo like this.
  16. Nibbels

    Nibbels Well-Known Member

    Joined:
    Dec 12, 2019
    Messages:
    165
    Likes Received:
    32
    To clear up the rumor about that temperature behind the SSR. I did heat the bed to 100°C for 2 hours now. Then I touched the L-aluminium I inserted and it was cold. It did not feel warm. <30°C I guess.

    For what I can tell there is no fire hazard ^^.
     
    Joe Pomo and yngndrw like this.

Share This Page