Telsa LDU HV cables' OD [SOLVED]

[asavage]

I've got a set of early LDU cables (1004872-00-B) on the way (via eBay), but until they arrive . . .

Do you have an accurate OD of these cables? I'm trying to order EMC shielded cable glands, and I need to verify that the HV cables for the LDU are <18mm OD, or larger than that. 18mm is the typical breakover point, where I order one part No., or another.

[later: I've added a "LDU HV Connectors" section to the LDU Wiki with some dimensions. Summary:
the outer insulation OD is ~17.3mm]

Typical early Model S LDU cables:

(The early Model S LDU HV cables have no bracket on the ends that insert into the GEN1 Rear HVJB, and that is the major distinguishing feature. As the Rear HVJB changed in late 2013 -- requiring a redesign of the HV cable assy -- there are a lot more later Model S HV cables available than early ones. I could have used either in my case -- I'm cutting off the ends that mate to the OEM HVJB -- but the best price available I found was for an early set.)

Typical EMC cable gland (for shielded cables, such as the LDU HV cables). Hummel 1.631.2500.50 HSK-M-EMC-D shown (currently USD$17 @ TME):

Here's a fun and short installation video by Hummel for this product. Hint: it's not the product shown below in the preview, which shows their "budget" product that requires straightening the shield wires and folding them back over an insert. The HSK-M-EMC-D line is a bit more expensive but much easier to install -- no fussing with the shield braid -- but is also longer/less compact.


By way of explanation, my particular project -- adding CCS DCFC to a 2014 Toyota RAV4 EV -- involves adding a pair of contactors in a new HVJB between the HV battery pack and what amounts to an early Tesla LDU. I'm cutting the OEM HV cables, adding EMC glands to maintain the cable shielding, and mounting contactors inside a die-cast aluminum IP67 box. Because the system I'm working with uses the early Telsa slip-in cable glands, and I have to continue to use them but lengthen the distance of the OEM cable run, it's just easiest for me to obtain an extra early Tesla HV cable set (with Tesla's proprietary glands) and use the new set on the LDU, in order to get everything long enough to connect to the new HVJB.

That's why I need to know the OD of Tesla's cables. It's also not convenient right now for me to measure my existing LDU cables. But, if nobody has a set laying about to measure accurately, I'll have the eBay ones in hand in a week or so, and I'll update the LDU Wiki with the info.

[Shockazulu1]

Instead of cutting the cables on your RAV4 EV.....
Why not connect a second set of cables directly to the battery, similar to the way they do it for the JDEMO add on kit?

[asavage]

(for those who aren't familiar with it: QCP's JDemo kit adds CHAdeMO DCFC capability to 2012-14 RAV4 EV and MB B250e, and it taps the traction battery pack by adding a "bathtub" to replace a cover that's on the OEM pack, then replaces cable bolts with studs to allow mounting extra cables to where the LDU cables would normally bolt to the HV pack. It's a bolt-on solution that doesn't cut anything except the plastic underbody cover (though it does defeat the HVIL switch on the OEM's cover, which is not replaced) and as a warranty-saving DCFC kit, it was a good balance of compromises. The RAV4 EVs are all long out of warranty now, though . . . )

Reasons NOT to use a QC Power JDemo-style "bathtub" on the front of the RAV4 EV traction battery:

* The JDemo method is clunky. It hangs low enough to require that awful bathtub/cable guard arrangement.

* Fashioning/fabricating a "bathtub" cover that is IP67 is not trivial; QCP's bathtub appears well done, but duplicating it wouldn't be a snap, and testing it for IP67 is not easy without having a spare pack with which to spray it with water at pressure. It's located just above the pavement and at the front of the pack (and the opposite end of where it's located on a Model S!) so testing for water ingress is required. IOW, fab'ing it is expensive in terms of time/labor.

* A new HVJB is required for the DCFC contactors anyway, regardless of how you tap the pack. (JDemo puts the contactors up on top of the stack.)

For those reasons, the cost of a set of used Model S LDU cables is minor, and it's much simpler to connect to the existing (cut) LDU leads than to fab and test my own "bathtub" adapter.

Now, if you happen to already own a JDemo "bathtub", then perhaps reusing those taps makes sense. QCP doesn't sell the "bathtub" separately from the (USD$3,000) kit, so unless you have a JDemo installation you have to start from scratch. Adapting off-the-shelf Model S LDU cables + die-cast contactors enclosure is a lot simpler (and cleaner I'd argue) than the "bathtub" adapter + extension studs route.

Here's a few pics of the "bathtub" installation (taken from the JDemo installation manual PDF):

[Shockazulu1]

Currently I don't have the JDEMO installation.
I'm just now beginning this process.
Right now you have the most info on what needs to be done, to do this project.
I see you don't have a patrion account, but I would like help you with your project.
I will be 3D designing the mount to hold the CCS1 connector in place.
I will be printing it with ABS so it will be able to hold up to heat.
Once I have one ready I'll pay for the shipping and send you one.
Thank you so much for your efforts.

[asavage]

I don't produce things in the way the Damien or Uwe or many others do; I mainly document things. I support uhi22 and Jack Bauer via Patreon, but I wouldn't dream of accepting donations that way, because I am not producing anything (other than Wiki articles, eg Tesla Model S GEN1 OBC, Tesla Model S GEN1 Rear HVJB). I've got ideas, I've thought about and discarded various roadmaps, but other than copy-n-pasting pictures, maintaining spreadsheets, and buying hardware, little actually has been accomplished that anyone could use.

Our RAV4 EVs are orphans, with under 2600 produced and many out of circulation (there were THREE RAV4 EV HV packs listed on CarPart in April!) the surviving herd is smaller every year. I am not planning to replace mine until it make no sense to keep it, and whereas in my previous life/commute I didn't need or want DCFC -- certainly not CHAdeMO, certainly not from QCP/Tony, and (less) certainly not for USD$3000 -- it's not an unreasonable price -- now I'm retired, living in a more rural area with longer travel distances and instead of buying, say, a VW ID.4 or Toyota bz4x to have similar cargo space and improved range, but also a hefty car payment, I'm more amenable to investing in what's in my driveway.

---

The CCS inlet port I have is from a BMW i3, and just holding it up in the fuel filler port area, it looks to be a tight-but-doable fit. The flap latch mechanism will likely have to go away, as I'll probably need that space. It will not surprise me to find I'll have to remove some of the sheet metal surrounding the current round gasket, squaring it up and removing the flap latch.

If you have a solution for mounting a CCS inlet -- ANYBODY'S -- to that space, I'm extremely interested. I wouldn't mind buying a different CCS1 inlet, if it fit the space better.

The cabling all runs down the front of the wheelwheel, outside the passenger compartment, and the new cabling should fit there nicely as well. I have not removed the inlet enough to take good measurements, but from a rough view there appears to be (possibly) enough room in the existing space to accommodate the CCS inlet.

If it occurs that a CCS inlet can't fit due to lack of depth, there's plenty of room inside to carve out an oblong hole, and from my cursory measurement it doesn't appear that this would entail modifying the plastic interior trim -- there is a little space (not much) between the side plastic panel and the metal behind it. The CCS' lock motor on the BMW part does not appear to fit well. How much metal will need to be removed will need investigation, and I've been sidelined by adding A/C to my residence

The following pics are with the interior side panel removed, when I added trailer/tow wiring and signal converter a couple of weeks ago:

[Shockazulu1]

What about possibly using a nacs Tesla connector that would fit the charger door opening more easily.....
Than use a CCS1 to nacs adapter each time we go to a charger?
Or even.... Since much of the Toyota RAV4 was based on the Tesla model s... Could we just source the missing fast charging parts from a wrecked model S?
I've attached photo of a fast charger held up close to our current charger port.

[Shockazulu1]

I wouldn't dream of accepting donations that way, because I am not producing anything

If I can send you the adapter bracket for the charging port opening, it would not actually be for payment, but for you to test it and provide feedback if changes need to be made. Would also want suggestions for it's initial development. True it is also another form of me letting you know your work is appreciated. Lol

[asavage]

NACS: it's not a standard. Now that the SAE is "expediting" a new NACS standard, J3400 . . .



. . . then perhaps in a couple of years it may be possible to semi-reliably DCFC non-Tesla vehicles having a NACS inlet port, but I can't wait that long for something I need reliable sooner.

Regarding CCS1 nozzle to NACS port: do they work? I haven't followed news on them, but I think I've heard that Tesla keeps moving the goalposts (ie updating firmware via OTA) to make the aftermarket ones unreliable -- or do I have that wrong? Simon is planning to use NACS, and he's acquired a Tesla inlet port to play with, but I'm unaware of a generic NACS inlet port; do they exist?

Without a NACS standard, I don't want to go there. I love the form factor of the NACS connector, for single-phase AC charging + DCFC it's a good design, and when we had a Model 3 I liked it a lot, but I don't think there's a clear path to using it for AC + DC in a DIY fashion . . . is there?

---

The OBC in the RAV4 EV and very early Model S is hardware identical, but the firmware used in the RAV4 will not do DCFC, it doesn't know how to talk to contactors nor negotiate PLC comms and we can't modify it. Nor will the Model S firmware work with the Gateway module in the RAV4, the Gateway rejects it. On top of that, there are no physical DCFC contactors on the RAV4 (as there is no equivalent to the Model S' Rear HVJB), and I also suspect that the DCFC comms on the early Model S are all CAN-based instead of CCS as are the later Model Ss, so it's really a no-go to try to interface.

There is no path to make that work presently, whereas what I'm planning to do can definitely be done (at least the way JDemo does it: put the car in "Ready" mode to get the BMS to turn on the HV rail, then pump HVDC to the battery via the LDU cables). I hope to improve on the JDemo experience by spoofing the OBC's CAN well enough to be able to fool the BMS to energize the HV rail without having the car in Ready mode, but even without that we know DCFC can be done on this vehicle, because JDemo does it.

Nobody has made a Tesla NACS port work well enough to use in a DIY project, AFAIK (I'd love to be wrong, though), and that's a hurdle too high for me to consider.

If you want to ditch the existing EV managment, you can roll your own by controlling the OBC yourself, talking to the BMS yourself, managing the DC-DC Converter yourself, and abandoning all the factory EV feedback on the IC and center console, but if you do that you may as well just buy another vehicle. I'm attempting to leave all that stuff undisturbed and just add on DCFC.

[asavage]

If I can send you the adapter bracket for the charging port opening,

What does it adapt? Got a screenshot of it?

[Shockazulu1]

What does it adapt? Got a screenshot of it?

If you remove the J1772 port on your RAV4, behind that is a bracket.
I intend to make a new bracket to replace that one.
I will 3d print the bracket with ABS plastic.
Now it's just a matter of both of us having the same CCS1 port.
I'll reference the photo you sent of the CCS1 connector that you already have.
I'm gonna start by purchasing the same CCS1 connector you have.
It appears they have 3 different after market versions of this connector, all with different mounting bolt locations.
And you have the BMW i3 connector port.

[Shockazulu1]

If you want to ditch the existing EV managment, you can roll your own by controlling the OBC yourself, talking to the BMS yourself, managing the DC-DC Converter yourself, and abandoning all the factory EV feedback on the IC and center console, but if you do that you may as well just buy another vehicle. I'm attempting to leave all that stuff undisturbed and just add on DCFC.

Before we look at installing the CCS1 connector where the J1772 connector currently is located.....
Would there be a benefit to leaving the J1772 alone, and leave it the way it is?
Like the JDEMO, we could put our DCFC port somewhere else.
I'm concerned with combining the wires for AC home charging with the wires for DCFC.

Attached is screenshot of a aftermarket connector I considered, if we still want to replace the J1772 connector.
This connector may be better than the one for the BMW i3, because it sticks out forward more.

[asavage]

I'm concerned with combining the wires for AC home charging with the wires for DCFC.

Tesla (NACS) shares the two wires for AC & DC charging. I can't speak to the later models, but the early Model S' OBC's input can accept (tolerate, really) either AC or DC. The Model S doesn't so much switch the two charge port wires' destination, as it just shorts the OBC's input to its output wires.

When AC is presented to the OBC, the OBC does its thing: converts the AC to higher/variable voltage DC.

When DC is presented to the OBC, the OBC activates the DCFC bypass contactors, connecting the OBC's input to its output. That's part of the clever bit Tesla did with NACS.

Our RAV4's AC wires are only 10 AWG and are suitable for ~40A AC; for 50kw charging -- what I'm planning for -- 125A DC are needed, which in turn dictates much larger leads, 1/0 AWG or 50mm2, so new leads must be run.

Since we didn't inherit Tesla's NACS nor NACS wiring, it's simplest to connect the existing 10 AWG AC lines at the existing J1772 port to the CCS1 port's AC leads (the top-half, the AC pins of the CCS1 are the same as J1772, physically & electrically). AC continues to feed the OBC just as it always has; DC gets routed to the new DCFC contactors. It's a CCS1 solution to a Tesla DIY EV, even though we are working with a semi-Tesla OEM build.

Before we look at installing the CCS1 connector where the J1772 connector currently is located.....
Would there be a benefit to leaving the J1772 alone, and leave it the way it is?
Like the JDEMO, we could put our DCFC port somewhere else.

Sure, we could put the CCS1 port elsewhere. I'm not interested in having the hood open for DCFC, as the JDemo requires, but perhaps a grill or front valance area? Let me know if you find a suitable location.

I don't know of a benefit of leaving the existing J1772 alone and adding a CCS1 port, other than not having to modify the fuel filler opening, but you're going to have to modify something to locate the CCS port. Modify the existing area, or a new area.

Attached is screenshot of a aftermarket connector I considered, if we still want to replace the J1772 connector.
This connector may be better than the one for the BMW i3, because it sticks out forward more.

BMW i3:

Is there an advantage to having the port project forward more? ATM, I don't know what space is available, and there's always going to be a trade-off: a port that projects forward still has to fit behind the fuel flap door, so it just moves the mounting flange back further. Put another way, the total distance front to back is likely to be the same.

Unless you're willing to leave the fuel flap off . . .

[asavage]

If I can send you the adapter bracket for the charging port opening . . . Would also want suggestions for it's initial development.
[ . . . ]
If you remove the J1772 port on your RAV4, behind that is a bracket.
I intend to make a new bracket to replace that one.
I will 3d print the bracket with ABS plastic.

This?

The inlet socket and some leads, G9081-0R01x (three different part Nos., because the 2012 at least didn't have a connector nearby and you had to drop the battery pack to replace a melted J1772 socket -- Thanks, Blink!) bolts to what appears to be a free-floating bit of plastic with molded flexible rubber around it, 58446-0R010, retained to the round hole in the fuel filler area by a circular metal retainer 77391-0R010.

I must be misunderstanding what your ABS bracket will attach to. The circular opening where that thing I will call a "grommet" attaches is too small to accommodate my CCS1's face. Some metal will need to be removed, I'm pretty sure.

Some of the metal bits back there:

These diagrams are maddeningly difficult to get; Toyota is putting completely unrelated diagrams next to correct part Nos., and even diagrams from completely different vehicles Third-party sites do much better, but watermark their work.

[Shockazulu1]

Hope to find a way that will not cut any metal.
I live where there is snow and they salt the roads.
Any cutting off metal will begin the rusting stage immediately.
Though I plan to store my RAV4 from first salting of the roads until the first spring showers can wash away all the salt from the roads.

[Shockazulu1]

Looking at the view of the charger socket area, it looks like everything is almost welded together.
May be difficult to create any space inside for the larger connector.
I'll make a attempt to take it apart when I get home from work today.

[asavage]

I'm reasonably certain that a CCS inlet will not fit within the current circular location. The bottom of the CCS connector is flat. Maybe modifying the top of the CCS port, eliminating the lock? Not recommended, but . . . what's your risk tolerance?

If you're not amenable to cutting metal at the fuel filler area, then you might look at installing at the grille area. Or try to figure out how to use NACS.

Looking at the view of the charger socket area, it looks like everything is almost welded together.
May be difficult to create any space inside for the larger connector.
I'll make a attempt to take it apart when I get home from work today.

Well, the metal parts are spot welded and seam sealed: typical construction. There are two fitment issues:

* Is there enough depth? Can the cables make the 90° turn forward? There exist CCS ports where the leads exit to the side; they can be ordered "leads left" or "leads right", but I've only seen them for commercial vehicles, and it doesn't appear to reduce the depth requirement.

TE Connectivity:
https://ie.rs-online.com/web/p/ev-charg ... 83524?gb=b

Something I've just come across that might fit better is something akin to the Amphenol HVCO series of CCS inlets, in which the lock motor location can be selected from various options:

https://ie.rs-online.com/web/p/ev-charg ... 41919?gb=b
Datasheet: https://docs.rs-online.com/6228/0900766b81638f58.pdf

One issue I see -- which is also an issue with other CCS inlet ports -- is that the rated current for the AC pins is only 32A; as you know, we can draw up to almost 40A. Surely a reason to go with a separate CCS port somewhere else, leaving the OEM one in place. In that case, you'd install a DC-only CCS port, which is an available option:

I don't have time to look for a comparable CCS1 version of this today. A 1-ph (CCS1) version is https://www.amphenolpcd.com.cn/products ... i-791.html, with 7 pins rather than 9.

I'll make a attempt to take it apart when I get home from work today.

You can't do much without pulling the wheel off, then removing the plastic wheel well harness cover. The picture I posted earlier of mine pulled out 2" is about as far as you can go from the topside, and you can't measure or really even see anything with only that much room. I can't pull the wheel off mine right now, so I didn't go any further.

[Shockazulu1]

I have the perfect idea....
We carefully remove the Toyota and EV logo on the front bumper.
We install the CCS1 connector in the same location.
We use the Toyota and EV logo to create a cap that will cover the CCS1 connector.
We will add a hinge to the cover so there is no chance the cover would get lost.
The biggest problem will be matching the paint. The TRI coat pearl white is very difficult to get right when painting with it.

[Shockazulu1]

If agreed with the front center connector with a flap covering it....
We could move to another possible option.....
Install a NACS port but use a adapter for CCS1.
We could keep the size of the door small.
When we charge we connect the adapter so we can use the CCS1 charger.

[Shockazulu1]

There is space in this bumper location for me to design a solid mount.
There is one metal strap/bar for a base. But I'll need to make metal bracing to stuffin up the strap/bar.
Once solid I'll add the 3d printed mounting block/plate for our connector to bolt onto. Gonna be sold enough to hold the charge cable firm even under all the weight of the heavy cord.

[Shockazulu1]

I'm now thinking, going the NACS Tesla connector route will just add more complications.
We are gonna have to put the CCS1 port directly on the front, and we will make a nice looking door to cover it up.
I'm thinking of mounting sideways with the bottom pointing twards the passenger side, so the cord will come from the same area where they charge Tesla (sort of). A Tesla charging port is on the rear driver's side and they back into the charging stations. We will pull straight in and use a charger on the front right.
Using this front location we can take care of this installation last.
We can do all our wiring, electronics and testing first before permanently mounting the connector to the front.
It's gonna take a bit of wire but not to much, to go around the cooling system that's mounted behind the bumper.
During our initial testing we will be able to temporarily mount the connector under the hood, and do the dreaded open the hood to plug in the charger.

[Shockazulu1]

So what we are left with now, if we move on to the next step.
1. Build the HV junction box.
2. With direct access to the HV input wires we will need a stand alone CCS control system.
(Hoping the car will activate it's own cooling system during the charging operation)
(I'm also hoping we will be able to run the a AC while charging, something that can't be done right now)
Currently what is the most reliable setup for the electronics we will need to complete this?
3. I will design the mount, and port cover, for installing the CCS1 connector in the front bumper.

[asavage]

At the simplest:

Doing things the JDemo way (placing the car in "Ready" mode), the BMS closes the HV pack contactors and activates the HV rail (LDU wires are "hot"). The BMS will then handle pack cooling/heating.

With the LDU wires hot, the new contactors in the new HVJB can connect the new CCS port to the HV battery.

For around USD$4k, ZeroEV has a standalone-ish CCS2 system. Not quite there for CCS1. And it won't interface with our BMS.

The BMW i3 LIM method is ready and working now.
* LIM module = $200 (used) - 450 (new + programming)
* LIM connectors or connector kit = ???
* HV sense board = $70-120
* IVT-S current/voltage sense module w/CAN = $250-350
* ZombieVerter VCU to control it = $350 (other controllers can be used, but it's more fiddling)
* Assy, customization, debugging.

The other CCS solution is QCA700x or AR7420, which is kind of working but not as "done" as the BMW i3 LIM setup. Johu & Celeron55 are successfully CCS charging using the AR7420 but there are lots of cases where it's not always working at a software level; the hardware seems nailed down, though.

Every CCS solution requires new DCFC contactors, which implies a new HVJB somewhere, but in our situation we don't need add'l fuses (they're at the front of the HV pack), disconnect (ditto) or any other bits: just a box with contactors & cable glands. And placement, of course.

None of the CCS solutions are bolt-on, plug-n-play, just hand over your money and go. And none are just plumbing and wires; there are software hurdles with all of them.

Then there's an uncharted path to getting the car to CCS DCFC by just plugging in the connector (ie car not in "Ready" mode, can be locked and can charge unattended, etc.) which is where Simon & I hope to get one day, and which is why there's a Wiki on the OBC now. This is the less simplest.

[Shockazulu1]

After work today I'm gonna run some simple security tests....
Does car stay in ready mode if I open the door?
If I lock it with remote?
Can car be put into gear if the key is not nearby?
I would like it to charge in ready mode, so I can run heat or air while waiting for car to charge.
But would like to walk away during charge and not have my car stolen.
But I could work around all that, if I could just fast charger that would be most important.
I wanna do this sooner rather that later.
Me and my car are only getting older. Lol

[asavage]

You can't lock the car in Ready mode unless the RFID FOB (the "key") is inside the vehicle. It's the same problem with DCFC with JDemo. That's why I need to do a CAN solution that emulates the OBC when DCFC.

Aside from that, it should CCS DCFC OK in Ready mode.

[Shockazulu1]

At the simplest:

Doing things the JDemo way (placing the car in "Ready" mode), the BMS closes the HV pack contactors and activates the HV rail (LDU wires are "hot"). The BMS will then handle pack cooling/heating.

With the LDU wires hot, the new contactors in the new HVJB can connect the new CCS port to the HV battery.


The BMW i3 LIM method is ready and working now.
* LIM module = $200 (used) - 450 (new + programming)
* LIM connectors or connector kit = ???
* HV sense board = $70-120
* IVT-S current/voltage sense module w/CAN = $250-350
* ZombieVerter VCU to control it = $350 (other controllers can be used, but it's more fiddling)
* Assy, customization, debugging.

Wow impressive. Great job gathering all that info.
Based on your research, I believe the obvious choice, if I want to get this done now...
Damien will get my vote for his solution.

I have found this link.
https://www.evcreate.com/ccs-fast-charging-in-diy/

The author has great prices for many of the items from your list.

I have worked with Toyota software hacking in the past, and feel I may find solutions to some or all of the security drawbacks we will have. We will still need to put in ready mode, but will be able to lock doors, and car would not be able to be operated without the key inside the vehicle.
It will probably involve connecting laptop with Toyota dealer software loaded onto it, to program through the ODB connector.