[WIP] Volvo Vee70 with Outlander Rear Motor

[crasbe]

Veeery nice! Udo will like that.
I think you have the book, right?

Maybe you can be somewhat compatible? But don't want to restrict you

I do have the book in print and digital form. Regarding compatibility I'll have to check if the pinout matches. BMW was very inconsitent with their ECU pinouts for the E46, even ECUs with the same connector have wildly different pinouts...

But thank you for the compliment. I'm sure Udo will like it, when the PCBs arrive and everything fits, I'll give one to him

[crasbe]

The PCBs for my ECU protoboard arrived 10 days early
Everything fits well so far, except for the locating tabs on the edge of the PCB. I changed the position of them before ordering the PCB, but apparently I still screwed it up. Oh well. Nothing a file can't fix.

I put on the STM32WB55 Nucleo to check if a Nucleo will fit on the PCB and it looks like it does. All other Nucleos are currently at work, that's why I used this one.

[crasbe]

A lot of stuff happened in the last month, unfortunately not as much on this project. But not nothing

First of all, a charging port of a BMW iX arrived (and a lot of high voltage cables), which has a nicely slim form factor to fit the limited space of the Volvo:

It is controlled via CAN and apparently has it's own CAN bus between the port and the charger. I did some limited work anaylsis but it only sends a message on 0x0161 when the lock switch wires are connected or disconnected, otherwise it's sleeping.
This probably needs logs to be reverse engineered.



Next, I was finally able to remove the clutch master cylinder (I found an 8mm plug made from Nylon, which is resistant to brake fluid).
The clutch master cylinder leaves behind a hole which has to be covered, so I designed a small plate, which will be 3D printed tomorrow and plasma cut from steel next week.
The stud you see in the picture also holds part of the pedal bracketry, so I don't want to rely on plastic for an important job like that.
Volvo had the so called "cover washer" for automatic cars, but it's no longer available and I'm not so keen on trying to find one on a scrap yard.
The finished cover will be published on Printables as usual.

The next big thing is power steering, which I finally began to make progress. The bracket sits on the left frame rail where the air box bracket used to sit. This is a convenient place close to the battery and adventageous for hose routing (at least I thought that).
Getting all the dimensions right took more than one try, but I don't want to go into toooo many details here. The process was pretty similar to the vacuum pump bracket design.

This is how the bracket sits in the car:

Originally I thought that I could just get some hydraulic hoses for the power steering, but the fitting type on the steering rack is not something that is standard in hydraulics unfortunately.

The plan now is to have the original lines modified. The company "KFZ-Leitungen Fabian Spiegler" repairs and modifies power steering lines, so I'll have them braze on fittings and the rest of the line will be made from steel braided hoses, somewhat like this:

The big project still left to do are the motor mounts and I have been procrastinating for far too long now...
A major help is the fact that Autodesk Inventor can (more or less) directly import the points I got from the Laser Tracker, so it's not toooooo hard to make the motor mounts... in theory at least...

[crasbe]

In the true spirit of "Don't get it running, get it right!" some small side projects were tackled today:

The first thing was the cover washer, which was printed in PLA for test fitting. It fits very nicely so next week it'll be plasma cut from steel.

The drivers side carpet got a good scrubbing so it can be installed soon (but I still have to think about how to patch the hole that's worn through).

The next project was swapping the drivers seat belt, which was frayed. This will not only mean a fail in the inspection but it is also seriously dangerous because the strength of the belt is reduced A LOT. Unfortunately this is a somewhat common problem and I could not find a single drivers side seat belt. HOWEVER it turns out that the drivers side seat belt and passenger side seat belt are identical and can just be swapped. The end is held in place by a steel pin. (Notice how the new belt is a lot cleaner as well?)

Then I wanted to install the door side marker light in the drivers side door... which led to another (unplanned) project. The silicone seals just crumbled away when I tried to insert the connector:

So of course I had to fire up Fusion 360 and design a replacement. The first prototype, printed from PLA, fit perfectly so I printed the final version in TPU.

The seals have a split, so they can be installed on the bullet connector without undoing the crimp. The split is then glued together with CA glue:

A little more information about the side project can be found at MVS: https://www.matthewsvolvosite.com/forum ... 7&start=10
And of course, the 3D model is published on Printables: https://www.printables.com/model/927466


All of these side projects might seem insignificant, but they were holding up the whole project (at least mentally for me). One thing depends on another which depends on another etc...

But I'll get back to "getting it running" soon.

[crasbe]

The cover washer including the gasket (in different flavors) is now on Printables as well: https://www.printables.com/model/836156

The first print in TPE was successful as well (the first one that doesn't look like a turd that is):

[crasbe]

In the true spirit of "don't get it running, get it right!" I wanted to complete the work on the carpet.

As you know, the carpet on the drivers side of my V70 had a hole worn through it (I still don't really know how or why, but I bought it like that).

Since carpetry is not really my strength, it took some time to figure out a solution that is in the realm of my capabilities. So first we start with a pattern:

This pattern is then made into a 2D sketch in Fusion. Furtunately it is really easy to import an image as a background, so that was quick work:

The decision was made to 3D print a patch. 3D printing is something I am familiar with, working with carpet isn't, so this promised to give the best result. The first prototype was printed from PLA, because it is much cheaper and much faster to print than flexible materials. The final patch is made from TPU with a Shore hardness of 95A. The prototype fit very well, so the final patch was printed the next day (it took nearly 5 hours instead of 1 hour).

To avoid ripping the carpet at the seam that is created at the stitching, a backing piece was 3D printed. The trick here is that the model was sliced without solid top and bottom layers, so only the infill grid remains. This allows for more flexibility (in a literal and figurative sense) during the stitching process:

Everything was aligned and held in place by a lot of pins and then I got to work:

I haven't stitched in a couple of years and even then I didn't have a huge amount of experience, but the results are not tooo bad I think:

Just the stitching took about five hours, the 3D prints took about seven hours in total and designing the patch took maybe an hour.
So now the carpet is finally ready to be put into the car again and I can finally reassemble the drivers side footwell

(Yes, I poked myself probably more than a hundred times and yes, my finger tips still hurt a little bit the day after. But there was very little blood.)

[crasbe]

Today the second part of the carpet voyage followed.
Unfortunately the rubber mat was hacked apart and worn as well before installing the carpet.

The previous owner decided to "fix" it with painters tape which obviously didn't really work. Therefore I got a roll of Siga Rissan, that is usually used to tape and seal moisture barrier foil. It sticks like crazy.

The center part of the rubber was crack-y as well, so it got a layer of Siga Rissan as well:

Now the footwell is nice again

I'm so happy to have this done. Now I can get back to making the car electric
(And the drivers seat is not in the floor of my flat anymore which definitely took waaaay too much space).

[crasbe]

After only *checks calender* four months I was able to beat the procrastination and get started on the motor mounts.
As you know, back in winter I took measurements of the bolt holes on the engine block and Outlander motor to design some motor mounts.
The Laser Tracker outputs these measurements as XYZ coordinates relative to it's own origin (you can set it to some other origin, but I didn't).

The initial plan was to write one or more fancy MATLAB scripts as I did with the gearbox measurements and do all the required transformations there. THIS was a big reason for the procrastination because the motor mounts are a lot less trivial than the gearbox bolt pattern. But last month I learned that Autodesk Inventor (as a spoiled university employee I have full access to it ) can import XYZ coordinates in a 3D sketch.
Why is that so significant? Now.. doing a projection of points is just a plaine (get it? plain-plane?) operation you can do with some clicks whereas in MATLAB you have to define the plane and do all the nasty linear algebra yourself.

I'll give a rough overview of the workflow, but to give a perspective to get to the end result you see at the end of this post, it took about 10 hours.

First we start with the imported points (the faint purple crosses):

What's not so easily visible in this screenshot is that each point is actually two points. The issue we faced with the gearbox as well is that the retroreflector sits a little bit recessed in the holes and therefore the position is slightly incorrect. The bigger the diameter of the hole, the deeper the retroreflector sits, the more wrong is the measurement. Therefore I measured the machined surfaces that are next to the holes with the retroreflector as well. From these points a plane can be created that the hole points can be projected onto, which negates the influence of the hole diameter. *)
In this screenshot you can see the reference surfaces and the sketches of the motor mount mockups:

The idea behind the motor mount mockups is that I treat the data as if the engine block sits in the center and I build "against" the original engine mounts.

These mockups are then 3D printed to verify that the bolt pattern matches the original mounts. It's quick and easy to do, so why not. The PLA I use will be recycled into new filament by the German Startup "RecyclingFabrik", so it's okay to print some more prototypes

With all the mockup motor mounts designed, the measurements can be set into relation. I always measured two mounts in one go, so they can be put into relation.

Since it is a bit hard to follow what's what, here are some (poorly) annotated screenshots to make everything a little bit clearer (hopefully):

But based on these fragments, I can start designing the actual motor mounts for the electric motor. The easiest one was the front motor mount, which can be bolted to the NEMA spacers:

The only issue I have now is that my trusty Sovol SV06 is a bit smöl for the NEMA spacers

The most challenging part will be getting the CV axle support bearing mounted rigidly, but that's a task for another day



*) Now that I am writing this, I realized that I made a very crucial mistake. The laser tracker does not measure the point of the surface the retroreflector touches but it the retroreflector's center. Therefore a 6.35mm offset is present in all measurements, but most importantly the points from which the projection plane is created. For the gearbox measurement, this was irrelevant because all points had this offset in the same direction, but now we have different directions... This is going to be... a lot of work to get it fixed..

[crasbe]

*) Now that I am writing this, I realized that I made a very crucial mistake. The laser tracker does not measure the point of the surface the retroreflector touches but it the retroreflector's center. Therefore a 6.35mm offset is present in all measurements, but most importantly the points from which the projection plane is created. For the gearbox measurement, this was irrelevant because all points had this offset in the same direction, but now we have different directions... This is going to be... a lot of work to get it fixed..

Furtunately it was easier than anticipated. With some helper axis all the mounts and elements could be moved by the 6.35mm into the right direction.
The CAD program looked a bit like a crime scene where the forensics team tries to reconstruct where the bullet scatter went

Unfortunately I already printed the bracket I was so proud of and it looks really nice, but it has to be modified and therefore printed again

This bracket is the one with the least amount of clearance to begin with, so the modified version has to be milled out. But it should be easy enough to do on the manual mill:

But first I have to go to bed, it's quite late. But without fixing that I could not have slept anyway

[crasbe]

A lot of things are happening right now, so it's time for a Sunday update

After the mishap with the dimensions, I wasn't 100% sure about it, so I printed this test piece to verify that the dimensions are right now. Looks good

Yesterday the plasma cut cover washer arrived. My partner has access to a CNC plasma table, so that is handy for small stuff like this. It only has to be drilled, tapped, painted and installed and then is the sub-project "clutch removal" is done

The next little project was printing the first of the NEMA spacer plates (it had to be cut to still fit on my printer )

Yesterday night I did the first (and second) iteration of the side engine mount. This is the first draft that I made to get an idea and to see how it might look like:

Turns out: This is not what it would look like
Tomdb gave me some very valuable hints and directions. The main issues with this are 1) that it'll be hard to precisely cut, position and weld the box section steel and 2) it is not really strong in the direction it has to be strong in. The biggest issue here is the CV axle support bearing, which (hopefully) won't experience any torque (if the bearing is not shot), but certainly some forces transfereed through the CV bearing.

The better and simpler approach is to use flat sheet metal, in my case 5mm steel. That's probably way too thicc, but since I don't want to do any fancy calculations with forces I don't know, it's better to build thicc than weak.

Placing the mount on my 3D printer was... challenging. It is pretty much maxed out and I had to cut the model a good bit (on top and on the left):

This was my longest print yet with the highest amount of filament used. In the end it took 19.5h and about 440g of filament. But the Sovol SV06 handled it like a champ and printed it without any flaws in PLA:

Of course, I had to test fit it to the motor and everything seemed to fit quite nicely. The only oversight was that I forgot about the sleeves in the CV axle bearing holder. Since I don't want to mount the CV axle yet, this wasn't a big issue though.

Since my partner took his engine hoise back with him again in February, I can't really get the gearbox and motor in the car. But I can get the subframe out of the car.
I wanted to do that anyway, because the subframe bushings are pretty much shot.

The engine mounts are pretty shot as well, but I only have the replacement for one of them. At this stage it'll be good enough, but when the ICE was still in, the gearbox was already bottoming out on the rubber endstop (which it should never touch). Especially in the last picture you can see just how much the mount sagged. The chain is usually very tight and you can not get it off the mount... yeah

Aaaaaaaaaand *drumroll*... here it is!
... in the next post. We reached the maximum amount of attachments
Cliffhanger

[crasbe]

Aaaaaaaaand here it is. Especially the front mount was quite a tight fit, but it looks good. It's just so nice to see everything coming together from the CAD system. And I'm even more impressed that everything held up, it's just PLA with not toooo much infill

The CV axle support bearing holder seems to sit in the right spot as well, but I didn't want to fiddle around with the drive shaft yet. It has a torn CV boot and is very greasy.

Unfortunately I probably won't be able to use my fancy phase gland (check my Printables page). The space is just too limited.

Now... the seasoned reader probably knows that I have a soft spot for fancy stuff, so of course I couldn't help myself and designed some feet for the subframe, so I don't have to drag it on the ground so much

[crasbe]

Remember when I talked about not trusting your measurements? Well... I made this fancy template and went ahead and not used it on the big printed piece...
On assembly everything was mostly fine, I was just wondering why everything fit rather tightly and used up all the slack from the original engine mounts.
Well, the reason is that I extruded the side engine mount plate to the wrong side. The engine block used to sit on the left edge of the blue part (the blue part is the mockup side engine mount) and I should have extruded the plate to the left, because the plate essentially is the engine block. However I did not do that and therefore everything was 5mm too long.

CAD says 524mm (don't mind the angle, it doesn't matter):

The real measurement of the engine block says 519mm. Exactly the 5mm that we added due to the extrusion.

MAYBE I'll print the big part again, but I'm not certain. 430g is a lot of filament and when I don't have to, I probably won't. (But knowing myself I probably will).

[johu]

Very systematic work
The motor looks so tiny on the subframe

[crasbe]

Very systematic work
The motor looks so tiny on the subframe

Thank you

Yes, the Outlander motor is quite a smol boi. But generally the same approach would work for the Nissan Leaf as well (which may or may not be planned for the future in another car (or in this one when the Outlander is too weak)).

[crasbe]

Thanks to the awesome work of larsrengersen, LRBen and Tomdb we now have control over the VAG MEB PTC air heater and Tom even made a 3D model (see his GrabCAD).

A rough check of the dimensions showed that it might be a good candidate to fit into the Volvo.
So I bought one from Poland for a reasonable price and removed the heat exchanger today.


Depending on who you ask this is a "dashboard out"-kind of job, but you can get around it. This is what we are talking about:

I struggled a lot to get the screws of the coolant line holding bracket out, because there is not much space. You probably don't have to, because you can just remove the whole block including the lines (which I found out afterwards):

But in the end it was out. You can "just" take out the whole lower part of the heater core box:

So now we are in an area I am more familiar with, making an adapter plate (now without a laser tracker ). First I took all the relevant dimensions and then turned them into a simple CAD model that can be 3D printed to verify the dimensions.

With the aforementioned CAD model of the heater, I designed an adapter plate that should allow fitting the heater to the Volvo quite easily. I will probably have to shave the guide rails of the heater to fit into the box.

The PTC heater will arrive in a few days and until then I'll have the adapters 3D printed and tested.

[crasbe]

So wow.. this was quite a journey again

First the adapter got printed:

Then I started to print the wonderful model from Tom, so I could safely test the heater core without breaking the actual thing (well and it hadn't arrived at that point):

The first fit was unsuccessful. The whole stack was just a bit higher than original and the black metal brace was in the way. What I did not know at this point is that you can just unbolt it quite easily...

Over night, my trusty Sovol SV06 printed the second half of the PTC heater, so we could do a faithful test of the fitment. What I already knew ahead of time is that the side "rails" of the PTC heater had to be shaved a little bit to fit into the original heater box, but that was no big job to do in CAD.

Then we went on to the next revision of the adapter, which got us a bit further than the previous one. And then the next one, which got es EVEN further.

In the meantime I got a bit sceptical whether or not the PTC element could possibly fit, so it was dryfitted. And look at that, it does fit. So it should be possible to get it in there... somehow..

The next adapter was going to the absolute maximum. 5mm thickness instead of 10mm to gain some space and the PTC moved as far to the back of the vehicle as possible. And would you look at that... it finally fits.

Now "all that's left to do" is to print the adapter from something more durable than PLA and install the actual PTC heater. Maybe I'll add a small fin to keep the two zone temperature control working (to some degree). And then of course it's electronics


But in the meantime I'll have this little fella to mount to the big 3D printed bracket. And guess what? That'll involvo some more 3D printing as well

[crasbe]

Friday we continued on our mission of "Don't get it running, get it right!" and replaced the headlight. As you can see, the reflector is very blind. That is the downside of the scandinavian cars which always have their low beams on. Increased security, decreased reflector lifetime (but to be fair, the reflector is 26 years old too). The actual reason for replacing it though was the broken mounting tab. I just didn't feel like investing a lot of time into fixing that (with fancy 3D prints) when the reflector is pretty much shot.

After removing the indicator (which will be replaced as well) I discovered a small rust spot, so of course this had to be addressed while we're at it. Since it is hidden behind the indicator, I didn't have to do a suuuuper clean job and just painted 2K acrylic paint on it:

The trunk lid received some temporary rust prevention as well. Ugly: yes, will most be hidden: yes, will it rust again: probably, but not so fast.

The next job was getting back to the AC compressor. First I created some plugs to protect the compressor from dirt and debris entering it (more than is already in it...). As usual, this is published on Printables as well: https://www.printables.com/model/949837

Getting the dimensions of the mounting flange right was a bit difficult because the mounting holes are recessed relative to the "belly" of the compressor. Therefore the first try was... not super successful
But hey, that's what 3D printing is for and the second try fit a lot better

Since the compressor will be mounted on the motor mounting bracket, it has to be designed as a part of that bracket of course:

And this is how it will look like on the bracket:

I wasn't supper happy with how close the bracket sits to to the little cap that protects the control cables, so the design had to be modified a bit:

Of course, the bolt pattern is published on Printables as well: https://www.printables.com/model/949824
Please check the models out and give them a Like for the statistics

I updated the wiki with some of the information gathered here to make it easier for people to use the Outlander compressor in their conversions: https://openinverter.org/wiki/Mitsubish ... Compressor

[crasbe]

MAYBE I'll print the big part again, but I'm not certain. 430g is a lot of filament and when I don't have to, I probably won't. (But knowing myself I probably will).

Well... I did reprint it. Initially I thought that the dimensions were incorrect again, but it turned out everything was right and the white piece just flexed a bit (a lot) under the load. Who would've thought?

Then I started to chop the piece up to prepare it for plasma cutting, but of course 3D printing prototypes ahead of time. While steel is not super expensive, 3D printed plastic is cheaper

And of course this was a good opportunity to test my new Prusa Mk4

This is how the puzzle looks. Please excuse the pictures, when you don't hold it right, it all falls apart spectacularly

For SOME reason not all dimensions in Inventor got updated and AGAIN we had misaligned holes... This was very unfortunate to discover at this point, because I did all the cutting in Fusion with the imported Inventor model. However you can't update the imported base model and it wasn't enough to shift the holes, the hole side plate had to be moved. Therefore my Fusion model is now a hot mess and the Inventor model is a hot mess too...

Of course everything had to be reprinted, but now it fits. The AC compressor holder got chopped as well, of course

And the next step is plasma cutting... but we'll have to wait till next week to see that
*CLIFFHANGER*!

[crasbe]

Just kidding, I couldn't wait until next week myself

Fortunately my partner has access to a plasma table at work, so I only had to buy the material. In this case, it's 5mm S355 steel. Probably overkill, but the idea is that there will be absolutely no discussion about whether or not it's strong enough

First we cut some test pieces to get an idea about the required tolerances. In the end we settled for 0.5mm clearance and 3mm circle size in the corners. That still requires some manual filing, but ensures a nice fit.

The first real piece was the AC compressor bracket because it is the smallest. The cuts are very clean (after cleaning the slag off), but a bit chamfered. This is to be expected for plasma cutting.

The big piece followed shortly after and everything fit together very nicely after a bit of filing.

Next we did some lathing. My partner did the short spacer and then I made the two longer spacers. The round stock is S355 as well and even though the cutting bit was quite shot, the result is pretty nice.

Furthermore, the adapter plates were ordered from Laserhub and should arrive in the next few days

[Markusvolvo]

Very interested in your Pcb/Ecu solution. Working on a volvo s70 -97 in Sweden, want to turn off the ABS light so I get working abs, please tell me more about your solution or if there is something you can perhaps build two of so I can buy that solution from you? Nice work indeed!

[crasbe]

Very interested in your Pcb/Ecu solution. Working on a volvo s70 -97 in Sweden, want to turn off the ABS light so I get working abs, please tell me more about your solution or if there is something you can perhaps build two of so I can buy that solution from you? Nice work indeed!

I'm curious to see more about your project, maybe you can open a Thread and share some pictures information?
Unfortunately, I didn't make much progress with the ECU, but maybe you can share what you've got so far. You should be able to read the ABS module without an ECU with a diagnostics tool and see what it is upset about.

[arber333]

Very interested in your Pcb/Ecu solution. Working on a volvo s70 -97 in Sweden, want to turn off the ABS light so I get working abs, please tell me more about your solution or if there is something you can perhaps build two of so I can buy that solution from you? Nice work indeed!

See here posts from Michael Neuweiler, he converted complex Volvo S80 back in 2016.

https://s80ev.blogspot.com/
https://github.com/neuweiler

I learned a lot from his posts.

[Markusvolvo]

Maybe would do that, post a thread on how I'm doing and how far I've come.

I use resolve-ev stuff through and through, even their ready-made cables, 30kwh nissan leaf battery is included. I don't use volvo's gearbox, I use nissan leaf's box, have made my own drive axles "cut volvo and welded nissan axles" so everything is ready there, working on the batteries now, there will be three boxes.

quick overview of where I am today^^

Will read the error codes on the Abs as you said and will get back to you What I have read is that the Abs wants to see that the engine is running for the light to go out, then assumed that the crankshaft sensor needs to act, therefore I am in the process of sending a pwm signal with 58/2 which should simulate the teeth that the flywheel has... Have sent pwm signal to the meter housing, then the tachometer moves but abs still doesn't go off. So I continue to work on a solution

[Romale]

Hm... i remember a guy who cut down the front part of the Nissan Juke engine and used it to fix the last bearing in place from both sides. In fact he used the OEM front bearing and flange part of the shaft since it was made for radial and angular loads anyway . Just a suggestion as you already have engine in pieces.

I did the same thing in one of my conversions, I used part of the engine as an adapter plate and this gave incredible precision to the articulation of the gearbox with the electric motor.