PDM Installation & Configuration: Step 6: Test Systems
Watch This Course
$149 USD $74.50 USD
-OR-
Or
8 easy payments of only $9.31
Instant access. Easy checkout. No fees.
Step 6: Test Systems
09.45
| 00:00 | - In this last step of our SR86 worked example here on configuring and setting up our PMUs from scratch, we've got to go through and test our work, we need to make sure that all of those output functions, those output channels are turning on and off in response to the functions that we've defined and that everything works correctly like we expected it to when we designed it. |
| 00:21 | Now it's more common in this scenario to find a problem with our initial thought pattern around how a function would work than something with the actual hardware being wrong. |
| 00:32 | It's really easy when you're designing functions to not think through every single case, it's always the edge cases with the bizarre out there combinations of inputs that you haven't quite thought of that can catch you up. |
| 00:43 | So in the testing that I've gone through in the SR86 here, I've come up with 3 particular problems that we're going to have to solve. |
| 00:52 | So I think it's best that we go through those in detail because they're really representative of the problems that you're also likely to strike when you're setting up PMUs as well so I think that'll be really valuable. |
| 01:04 | So the first of these I bought upon myself and it was absolutely a case of not thinking through those edge cases. |
| 01:10 | If we pop over to my laptop quickly we can see I'm back connected to PMU1 and we're going to have a look at our fans override signal here. |
| 01:20 | So I can command our fans to be overridden on via the button on our keypad here. |
| 01:26 | So that would be our K fans override input. |
| 01:29 | That's going to turn our fans on, turn that back off again but they're also overridden by our pit mode input so that's our button on the steering wheel there. |
| 01:38 | Now I thought it would be nice because on the keypad when I press that fan override button, it changes to being backlit green, telling us that that output is on, we've pushed that button and it is overriding the fans to be on. |
| 01:53 | And I thought hey it'd be really nice if that flashed up green when we push the pit mode button as well, just giving us a good visual indicator that the fans are being overridden and driven as well. |
| 02:03 | So I thought what I'd be able to do is come into my keypad setup here, my fan override button and I'd be able to define one of these set channel fields down here. |
| 02:16 | So there you can link to another element in the PMU and it has sort of the same effect as pressing the button on the keypad. |
| 02:24 | So I could have had our set channel here, I thought if I linked that to my pit mode, we'll accept that and accept everything through, then if we're watching our fans override input here but I'll press the pit mode button, then absolutely yes it changes to 1, pit mode is also 1 because we've pressed that button, so that means it has lit up green on the back of our keypad here. |
| 02:54 | Exactly the functionality we want but the edge case is now that when I turn our pit mode off, our coolant fan override input from the keypad actually remains on and that's probably not going to be expected behaviour, it'd be really easy for the driver to be exiting the pits, get to the line where he can accelerate, disengage pit mode and just go hell for leather, oncoming traffic, there's a lot going on, you've got to watch out and not realise that the fans are actually still overriden on. |
| 03:23 | So probably not exactly desirable behaviour there so bit of an edge case that I didn't think through means that input that I wanted to set up there isn't quite going to work so I'm going to go through and actually just get rid of that one. |
| 03:36 | I'm not saying there isn't a way to set this up like we would want it but absolutely there'd be a bit more thought involved in that functional design and it's best to have simpler but more coarse functionality, it's easier for the driver and everyone to understand than something that's more advanced but possibly prone to edge cases causing the issues. |
| 04:00 | So with all that disabled I can accept that and we'll just double check that everything's correct again and so our fan override is off, turn our pit mode on, that's going to turn our fans on 'cause it's an "or" function so they are active but it hasn't changed our fans override input. |
| 04:18 | It's not lighting up green, it's not ideal but this is going to be a simpler overall functionality for us to understand so testing procedure there to be honest it actually wasn't me that picked that one up and this is something I mentioned in the course material that you want to get somebody else to jump into the car and press all the buttons. |
| 04:35 | Just press them, just go hell for leather and make sure that nothing unexpected happens so this was one of the other guys I had jump in here and go hey just cycle through some of the inputs and see what's happening and see if you see anything unexpected and that was one of the elements that came up. |
| 04:49 | Now the second of the 3 important fixes we're going to make is that, the last 2 fixes are related to one another but they're sort of the opposite sides of the same coin. |
| 05:02 | So our water pump, we're PWM controlling and on the surface it looks like we'd absolutely be capable of doing that with our PMU, it's got a specified 9A draw and we've got it hooked to a PMU output channel that's 25A capable so that's lots of head room there but as soon as we've bought PWM into the equation, I was actually seeing quite a lot of heat build up in our PMU and that's likely to cause us an issue when everything's loaded up out at the track. |
| 05:31 | So the way I've got around that issue is to actually parallel the output on that one. |
| 05:36 | So I'll just rearrange here for a second and connect to our second PMU and show you what we've done there. |
| 05:43 | So I've got our second PMU connected here now and we've got that heat build up problem and the way we're going to get around that is just to parallel that output through another transistor, spread that heat load around. |
| 05:56 | So our water pump output here, we're going to change the pin to be a double, so that's going to set up a parallel output and the other channel I'm going to choose is going to be output channel 15 there. |
| 06:05 | And the reason for that is if I just scroll sideways you can see the physical location of the transistors which are our heat geneating elements, output one here is on one side of the PMU and output 15 is on the other so that's going to spread that heat load around and it'll be dissipated as easily as possible. |
| 06:26 | So I can accept that there, so pretty simple fix on that one to be honest, really common fix for PWM applications. |
| 06:36 | Can't really overstate how important it is when we're running things with PWM to really test that output 'cause heat generation, it can be a lot more severe than you think it will, particularly through just a single transistor. |
| 06:51 | So having to double things up like that, even though it looked like we had lots of headroom on just a single output channel, really not an uncommon situation. |
| 06:59 | Now the third problem that we discovered that we had to fix is actually a little bit of an uncommon situation, it's not one that I'd struck before and I had to consult with Ecumaster about this one to get to the bottom of what was really going on. |
| 07:10 | And it was to do with our compressor output here, so we had that on output two with an in rush current of 100A set at a maximum output of 23A. |
| 07:20 | Now the compressor when fully loaded up will draw 23A so that's very close to that max current limit of 25A but we weren't actually having that. |
| 07:30 | Even when we were turning the compressor on, when the system already had quite a lot of pressure. |
| 07:36 | Where we were seeing an issue was when we were turning the compressor off. |
| 07:40 | Which was quite unusual, well it's not unusual really because of the way electromotive loads work, there's a lot of current passing through an inductor there. |
| 07:48 | So you go to switch it off and that inductor resists the changes in current which can generate really large voltages and that's exactly what was happening, it was throwing an over voltage error in our PMU when we were trying to turn the compressor off. |
| 08:00 | When things got a little bit hot and under a lot of load was when we were seeing that. |
| 08:04 | Now the fix is exactly the same as for our water pump there. |
| 08:11 | It's sort of the opposite side, we're not talking about PWM controls supplying power anymore, we're talking about being able to stop the supply of power in the timeframe that we want. |
| 08:19 | So what I've done is simply parallel that up, our compressor output here double and similar story as before, we're going to spread that load around and we're going to put that on output 14 as well. |
| 08:32 | Everything else is the same, all the driving functions for that are the same but now we've got that paralleled across 2 outputs. |
| 08:38 | Now that has meant that we've had to modify our wiring harness here. |
| 08:42 | Now the wiring for these devices absolutely up to the task of passing the available current. |
| 08:48 | The problem that we've struck is heat generation within the PMU so what we've done to avoid needing to run extra wiring through our already constructed harnesses, is we've bridged that together at the point where it actually plugs into the PMU. |
| 09:03 | Because once again those wires that are running to the compressor down here and the coolant pump in the engine bay are absolutely capable of passing the amount of current, our current limit settings are totally fine around the size of those wires, but it's just the PMUs having trouble dissipating that heat and dealing with that voltage spike. |
| 09:21 | So that's been a really good solution to those problems, good easy fixes there, getting us just around those issues. |
| 09:29 | So hopefully that's a really good representative case study or example of 3 situations that we struck when we were undergoing the testing procedure for the worked example in our SR86 racecar here. |
