Electrical System Detailed Design Continued
As laid out in a previous electrical post, the electrical design was broken down into five sections. 1-3 were discussed in detail in that post, while 4-5 will be covered here.
- Major Component Selection
- Circuit Protection
- Safety
- Minor Component Selection. The electrical schematic can be detailed, and auxiliary components added: terminal blocks, relays, cable passthroughs, etc. Connections to inputs and outputs, the control panel, and cabling can all be specified.
- Heat Calculations, Cooling: The heat generated by each component can be used to determine if cooling will be required for the control panel. If a fan or air conditioner is required, this will need to be added to the electrical schematic.
- Electrical Cabinet Layout: With all the components selected, a layout can be made. Often, a rough layout is imagined earlier in the design (to put a control panel on order), but without all the above steps complete, a finalized layout is unlikely, so waiting on layout until the end is recommended.
Minor Component Selection and Schematic Detailing
Now that all of the major components were selected, the schematic could be detailed, and all of the minor components selected. It was a challenge to fit everything into the three-page limit for Design Spark Electrical, but with some work I fit it all in! Below is a preview of the power page, the initial release is in GitHub.
Terminal Blocks and Relays
Most of the controller inputs will be wired as two-wire signals (with the only exception being the limit and homing switches). The controller itself only has unique terminals for the inputs, then has a single 24VDC and 0VDC terminal. For ease of wiring, it was decided to have a multiconductor cable go from the controller panel to the main control box and have that single cable land into a terminal strip that will break out to all the IO cables and distribute the 0V.
For the limit switches that are three-wire, a three-level terminal strip will be used. There are special terminals meant for IO that have more compact distribution, but the cost and availability of these was prohibitive for this project compared to generic 2 and 3 level terminal strips. The benefit of these styles of terminals is that commonly used potentials (24V and 0V) can be distributed to multiple adjacent terminals using jumper bars instead of having to daisy-chain the connections with lots of short wires.
Outputs from the controller will also be mostly two-wire with some that go to additional relays. The outputs will land to multi-level terminal blocks just like the inputs, and additional din rail mounted relays will be used to turn on/off outputs that will require a larger current to switch.
To keep costs down, bulk terminal blocks and a multi-pack of relays were ordered.
Cables
When trying to keep costs low, it is important to plan out all the detailed components to cut down on number of orders, shipping costs and get quantity discounts where applicable. Flexible cables were used for cables that would go through the drag chains, while cheaper PVC jacketed cables were used for stationary locations.
For all the limit and home switches, I wanted to have connectors close to the sensor so that if there was a damaged sensor (or cable) it could be replaced without having to replace the other. The Panasonic slot sensors for homing were bought with a connector on the sensor, so the mating cable was bought for these.
For the proximity limit switches, they come with a pigtail of cable, so it was much more open ended. I really like M8/M12 connectors as they are easy to assemble, well-sealed, fairly inexpensive and most importantly very durable. Both M8 and M12 cables are typically rated for at least 3A (usually 4A) and my circuit protection for the IO is all at 3A. M8 connectors and cables are a little smaller and were slightly less expensive, so that’s what was selected for this project. Sometimes the M8 connectors can be harder to assembly (due to their small size), but I specifically selected screw terminal connectors that should be fairly easy to install. Starting at the sensor, there will be a small length of cable that comes from the sensor (cut short with a small service loop), then a M8 male field wireable connector will be added to the cable, then there will be a pre-molded M8 female connector cable that runs all the way back to the control cabinet and lands directly into the 3-level terminal block.
For passing signals between the HMI and control cabinet, bulk multi-conductor cables were ordered. A 25-conductor cable for all of the inputs, 12-conductor cable for the outputs, a 5-conductor cable for the power and ground, 4 conductor twisted-pair cable for the stepper drive signals and a 7-conductor shielded cable for the VFD signals. Using cables broken down by function will allow for some really clean wiring and will help to keep the cabinet tidy with a cost that is pretty similar to (and possibly even less than) bulk wire. These cables were sourced between Igus and Automation direct based on cost.
Cable Entry
To pass all the cables into the cabinet coming from different areas on the machine, a bulk passthrough was purchased to make for quick and easy install.
The power cable passes into the cable separately with a cord grip so that it can be isolated from all the IO. The cable that goes from the VFD to the spindle is similarly isolated and passes through a cord grip with the same part number as that for the power cable.
Separately, there is a tube that connects the HMI to the control cabinet for the bulk cables that connect these boxes. A hole from this tube into each box was made to run the cables.
Din Rail and Duct, Wire, Ferrules, etc
Bulk din rail and duct was bought for this project along with a lot of other “consumables” (ferrules, ring terminals, wire labels, zip ties, etc). Since this panel is very crowded, 40mm wide duct was used. This should be sufficient for most locations, as there isn’t too much internal wiring. The only area that will be tight is where the external cables for the sensors come into the cabinet and need to make their way to terminal blocks. Since these are all cables, if some need to be routed around the duct, I won’t be too sad.
The din rail is all standard 35mmx15mm. The deeper 15mm rail was chosen over 7.5mm as it can be easier to get components on and off and if we ever want to put Beckhoff components on our spare din rail, we will need 15mm.
For the wires, small spools/quantities of wire were ordered off McMaster. Surprisingly, when looking at the 10,14 and 18 AWG wires, the McMaster per foot prices were very competitive for small quantities. If I was a panel shop or building more panels, then getting larger spools off Automation Direct or another site would be cheaper. For each size, I estimated lengths for each color required (blue, white-blue, green, black) and just bought that amount of wire. All wires (and cables) purchased for this project are rated to at least 300V. The wires are all stranded, and ferrules/ring terminals will be used for all terminations.
Ferrules, wire labels, ring terminals, zip ties, etc. were bought in small bulk orders.
Heat Calculations and Cooling
Since all the major components were selected already, it was easy to tally up the heat generators and determine that a cooling fan would be necessary to keep the panel temperature to a manageable level. Rittal has a good online tool to calculate the actual required airflow, so that’s what I used to run the numbers.
Inputs and Assumptions:
- 620W of heat to dissipate
- 150W power supplies: 20.3W x 2 = 40.6W
- 600W power supplies: 84W x 3 = 252W
- Stepper Drives: 13Wx5 = 65W
- VFD: 261W
- 600mmx600mmx300mm cabinet that is mounted to allow for free convection on all sides (dissipates ~62W, calc has a 250mm depth cabinet as Rittal doesn’t have this exact size – conservative estimate)
- Max temp of garage: 30C with altitude of 250-499m
- Max temp internal allowed 40C (VFD limits this)
With the listed setup, the calculator shows that climate control is required and with an air throughput of 184 m3/hr.
Based on this analysis a 24VDC fan was added to the cabinet. To meet the required airflow, a 124 cfm fan was selected (~210 m3/hr). For both the fan inlet and outlet, filters will be added to try to keep dust from entering the cabinet. For this project, the fan will pull air into the cabinet in an attempt to have positive pressure inside so that the entry point for all air will be through the fan’s filter.
Thermostat
Since this will be a dusty environment, we only want the fan to run when its needed. A small thermostat was added in series with the fan so that it will only run when the cabinet is getting hot. This small component price will definitely pay for itself by increasing the longevity of the filters.
Control Panel Layout
The layout for this panel was quite the spatial puzzle. Luckily, we have an expert “shapes engineer” that helped brainstorm ideas. We wanted to make sure to allow airflow past the stepper drives and wanted to try to get the VFD inside the panel to keep it protected from dust.
To help visualize different options, the major components were laid out on the back panel and some strips of paper were cut to show the duct locations. Through this exercise, we realized that if we mounted the VFD and 600W power supplies to the sides of the enclosure, we would be able to fit everything without restricting airflow too much (since the enclosure is very deep for its size).
Once our layout was selected, the components were all imported into SolidWorks and arranged in the 3D model to ensure fit and make drilling templates.
The 3D arrangement was especially important because of the way the fans, power supplies, and VFD mount to the control box.
