Lexus GS450H VCU Support Thread

[xp677]

It works on GWS191 only, so 2006-2011

[Gigas]

I got mine in today for only 400$ came with full wire harness, transmission mount and the oil pump! now to get the board and figure this thing out. I couldn't get an image to load so I just linked my Instagram

[Dan--T]

Hello, I'm planning my first EV conversion and intend to use the GS450H box and Damiens VCU board. I've bought a crash damaged Lexus and next step is to get the board and get the car to move under EV power using the existing battery.

Could somebody help me with a few questions?
https://evbmw.com/index.php/evbmw-websh ... 0h-vcm-kit

1. The V2 boards currently on sale - How many components require soldering on them? Are all of the Surface mount parts fitted? I am fine soldering components but surface mount parts I have never done

2. Is there a built and fully tested board ever planned to be on sale from EVBMW? - If so, I would potentially wait to purchase one of these

3. Currently the description says V2, but the product code is gs450h_v1_built - Could anyone confirm if the current boards on sale are definitely V2 not V1?

4. Is this board designed to fit inside one of the existing OEM casings? Or is a separate enclosure required? does it connect to the OEM wiring harness or is a link harness required?


Thank you in advance, looking forward to learning more about on this new EV world!
Dan

[Jack Bauer]

1)all smd parts are placed.
2)No
3)all boards are now v2
4)Please read the wiki

[Dan--T]

Thank you!

Very clear, and the Wiki is extremely useful, sorry should have re-checked that!

[Gigas]

Thank you!

Very clear, and the Wiki is extremely useful, sorry should have re-checked that!

I would recommend taking a couple hour and reading this entire thread as it answers all your question plus your learn a ton. A pinout is in there too the box fits in a mouser part that is in the BOM and can be had for 12ish bucks

[Dan--T]

Thanks yes I'm going to go through it all in detail now, got a lot of learning to do!

Just placed my order for a V2 kit, so looking forward to getting things moving in the new year!

[Gigas]

Thanks yes I'm going to go through it all in detail now, got a lot of learning to do!

Just placed my order for a V2 kit, so looking forward to getting things moving in the new year!

It's a lot of fun learning - I have a thread with a couple of the things that I don't quite understand but it really fun figuring it out. I'm glad this forum is here us learn

[ggeter]

To make it more clear, i have made a complete wiring diagram for the V2 board in visio.

As i am also missing the original wiring loom, unfortunately i can not check if all connections are correct.

I would really appreciate if everybody gives there remarks so i can modify and finalize the drawing.

Please also find the Visio file here:
https://www.dropbox.com/s/e42971nsftjav ... .vsdx?dl=0

Is this PDF your most current version of the wiring diagram? Or, do you keep up the Visio?

[Paul_subi]

Hello all
I am new to the forum but I was watching many of the project's for last year. On january I bought a bare V1 board from Damien,install all the parts, menage to "get it working" . Over the last year I had some up and downs in my life, finaly got time to get back to the project. At the moment I cannot get it to spin really fast and it is behave like that :



Gearbox is starting to spin, accelerate but suddenly it is stopping really hard, amp meter goes into really high value.
I did try to run it on higher voltage (150V from lev46 battery's) and it is the same. It looks like some kind of problem with resolver but i did check it few times and it looks like it is connect correctly as it is stated in the scheme provided in the topic. Please help

[windydrew]

I've had my Smith Newton for going on 2 years now, just sitting here and I find myself wanting to get it running. I've decided I'm going to pull the gs450h to power the Smith. I also have a 22kW diesel genset that will give me extended range and site power while constructing Solar Projects here in rural Kansas. What will it take to get my 450H controller board that I ordered? Just looking for a timeline. I'll be getting the genset mounted and 240v G1 Volt pack mounted while I wait. I'm pretty excited as I have been dreaming about this for a while.

[Dilbert]

It's glued/bonded in - I had to smash mine out. I think I used a screwdriver and then a craft knife to clean up the rest. Be careful not to cut yourself or scratch the aluminium, it's a mating face for my cover!

DC bus cover pictured in my last post is done. I just want to add o-ring grooves around the cable entries, and a million other tiny things that I'll notice.




Now onto the converter input cover!

Does anyone have a Step/Iges/SLDPRT file for the GS450H inverter that they are willing to share? I've access to a 3D printer at the moment, so i would like to print one.

[Gigas]

It's glued/bonded in - I had to smash mine out. I think I used a screwdriver and then a craft knife to clean up the rest. Be careful not to cut yourself or scratch the aluminium, it's a mating face for my cover!

DC bus cover pictured in my last post is done. I just want to add o-ring grooves around the cable entries, and a million other tiny things that I'll notice.




Now onto the converter input cover!

Does anyone have a Step/Iges/SLDPRT file for the GS450H inverter that they are willing to share? I've access to a 3D printer at the moment, so i would like to print one.

If I had more detail Dimensions I would make it - but when I get to that point I will design one but I don't have the copper studs or wire to make a correct render in fusion

[xp677]

Uhh yeah I have the files for this but am away from home right now. IIRC they weren't completely correct, I needed to make a few changes to get them to fit and even then it wasn't great. I need to dig the unit out at some point and finish them.

I was actually thinking of selling these parts since I lost my job due to COVID, I never got round to it though!

[bobbyj_408]

So I want to make sure I understand the precharge logic in the VCU. When the inverter is precharging, the inverter will report back the dc bus voltage. As the voltage is increasing, that will get read in to the VCU. Once it has reached a threshold, it will close the main contactor. In essense, the inverter is the one sensing the voltage, the VCU receives that value, and controls the contractor.

On top of that, I notice that the threshold value is hard coded to 400. What if my battery isn't at 400V? I'm using a Chevy Volt pack and I believe it's nominally at 350V. What that also means is that when it's low on a charge, it can be quite a bit lower than 350V, and when it's on a full chaege it can be higher than 350V. How can we compensate for this? Would the idea be to set the threshold very low? If I do that, and the pack is full, wouldn't that cause a current rush?

[Jack Bauer]

Just change the value to suit your pack.

[nkiernan]

Has anyone had any issues with MG2 rpm not showing on the VCU wifi link!?

I've tried two VCU boards, on both a wifi connection to the phone and a laptop, and on two separate gearboxes, but the MG2 rpm stays at 9000. Not sure where its getting that figure from as there's no 9000 throughout the VCU code.

All other dials are working as expected. Even when I manually send test values over the wifi connection, all others work but not MG2! Looks as if the VCU URL just isn't getting the correct variable? Based on Damian's v7 code.

[Jack Bauer]

The gauges default to 9000 if no data for that item is received from the mcu.

[nkiernan]

When testing the wifi connection, I tweaked the code in the wifi function as shown:

Serial2.print("v111,i222,p33,m3333,n4444,o22,r33,q55*"); // Test string

Image below shows the result. As before, mg2 rpm is not registering. I would have expected if there was a wifi issue, the other 7 values wouldn't work either!? Its as if the identifier 'n' for mg2 rpm is not registering with the GS450H wifi dashboard. Is it possible I could have messed up the dashboard, doesn't look like it?

[xp677]

So I want to make sure I understand the precharge logic in the VCU. When the inverter is precharging, the inverter will report back the dc bus voltage. As the voltage is increasing, that will get read in to the VCU. Once it has reached a threshold, it will close the main contactor. In essense, the inverter is the one sensing the voltage, the VCU receives that value, and controls the contractor.

On top of that, I notice that the threshold value is hard coded to 400. What if my battery isn't at 400V? I'm using a Chevy Volt pack and I believe it's nominally at 350V. What that also means is that when it's low on a charge, it can be quite a bit lower than 350V, and when it's on a full chaege it can be higher than 350V. How can we compensate for this? Would the idea be to set the threshold very low? If I do that, and the pack is full, wouldn't that cause a current rush?

Ideally you want to compare the reading from the inverter with the reported pack voltage from the BMS. So that as the battery pack voltage decreases, your expected dc bus voltage changes accordingly.

For your system, it might be an idea to grab the pack voltage, and set the threshold based on this during startup.

For my system, I sum all the cell voltages to get the pack voltage, and my conditions to close the main contactor during precharge are:

1. inverter DC bus voltage is within x of the pack voltage
2. current through the battery pack's current transformer is less than y
3. more than z milliseconds has passed.

I don't have the values to hand, but it's probably something like 10V difference less than 5A, and 1 second passed.

I know that Damien and others recommend a voltage/current sensor which could be useful for this as well.

From experience, the inverter's readout is not especially accurate.

The point of precharge is to prevent an inrush when applying pack voltage to the DC bus capacitor within the inverter. In reality, measuring the current flow into the inverter is a more direct way to handle this.

[bobbyj_408]

Ideally you want to compare the reading from the inverter with the reported pack voltage from the BMS. So that as the battery pack voltage decreases, your expected dc bus voltage changes accordingly.

For your system, it might be an idea to grab the pack voltage, and set the threshold based on this during startup.

For my system, I sum all the cell voltages to get the pack voltage, and my conditions to close the main contactor during precharge are:

1. inverter DC bus voltage is within x of the pack voltage
2. current through the battery pack's current transformer is less than y
3. more than z milliseconds has passed.

I don't have the values to hand, but it's probably something like 10V difference less than 5A, and 1 second passed.

I know that Damien and others recommend a voltage/current sensor which could be useful for this as well.

From experience, the inverter's readout is not especially accurate.

The point of precharge is to prevent an inrush when applying pack voltage to the DC bus capacitor within the inverter. In reality, measuring the current flow into the inverter is a more direct way to handle this.

Thanks xp677!

Yeah this is more along the lines of what I expected. I fully agree that it's probably best to set the threshold based on the pack voltage at startup. I think I might try what you summed up in 1 - 3 just to cover all my basis. Looks like the next thing I need to purchase is a shunt to measure the current. Probably a good thing to have anyways.

[celeron55]

I figured you don't really need to know the current, because you know that there's no excess current if the system voltage is able to reach a sufficiently high value when fed through the precharge resistor.

I also think it's important to limit the timing of the precharge to not take too little and not too long. If it takes an unpredicted time, you need to check what's wrong.

[xp677]

Yep, the reason I base it on current is because my current readings are from a dedicated sensor (CAB300), the update rate of this sensor is higher, and the resolution is greater than relying on the inverter voltage readout.

It's the inrush current which we are trying to limit, so as long as that is within spec for the contactor (I'm using EV200 contactors), all should be fine. I have a min precharge time of around a second, it might be less (like 500ms), and a max of 3 seconds before the BMS aborts.

[ggeter]

I think I'll be using one of these in my setup and not using the VCU functionality. Measures delta V of 10 or less before engaging main contactor. Plug and play. Any thoughts?
https://www.evwest.com/catalog/product_ ... chk1c6rd25

[xp677]

The VCU already does this so it's $58 that you don't need to spend, but it's your money so if it makes you feel better then go for it.