UMC Wiring

[W.Petefish]

Does anyone know how the current revision of the UMC detects the current available from the adaptor?

I suspect that it has to do with a resistor, but what else?

[TEG]

Related threads:
how do the cables work technically? (http://www.teslamotorsclub.com/showthread.php/5424-how-do-the-cables-work-technically)
Adapter for UMC 32A wanted for conversion to Mennekes plug (http://www.teslamotorsclub.com/showthread.php/5342-Adapter-for-UMC-32A-wanted-for-conversion-to-Mennekes-plug)

[W.Petefish]

That was a bust.

I was talking about the, just for example, 10-30 end to the California connector. THESE (http://shop.teslamotors.com/collections/charging/products/universal-mobile-connector-adapters).

[S-2000 Roadster]

The 'manual' for the UMC shows that there are conductive metal rings on the changeable part of the adapter. The rings vary in size, and evidently make contact with an array of conductors. There appears to be a unique ring shape, size, and placement for each possible current rating. Since the various physical plugs on the ends of the adapters each have a standard maximum current rating, Tesla Motors simply builds the adapter cables to match the standard maximum current rating with the appropriate metal ring. I am not aware of anyone else doing this, so I doubt that you can buy those rings anywhere else.

So, basically, it doesn't literally detect the available current (that would be impossible anyway), but rather it is 'keyed' with a conductive metal key that corresponds to the standard maximum current rating of the attached physical plug standard.

The bigger, 'smart' end of the UMC has all the brains to detect the metal rings and 'know' which physical plug is attached. The brain then generates the correct electronic pilot signal for the Roadster. In effect, the bump in the cable converts between physical keying and electronic signaling.

Note: There might not actually be any resistors in this part of the keying. There are certainly resistors in other subsystems of the charger, but they're not necessary for this part (unless they use resistors to reduce the number of input pins on the embedded processor).

P.S. If a 'standard' receptacle is improperly wired to the wrong breaker, the UMC will still announce the assumed maximum current, not the actual maximum current.

[TEG]

I know how the old RFMC does it (pre-dating the Tesla UMC), but I would assume the Tesla UMC does something similar with different resistors and/or diodes to let the "brains" know what kind of plug end you have attached, and it will then send an appropriate pilot signal to the Roadster. "Keying" by shape of the metal ring seems far fetched to me.

[S-2000 Roadster]

"Keying" by shape of the metal ring seems far fetched to me.
This could be another one of those technical details of the Roadster explained to me by my salesman which turned out to be way off from the actual technology. It sure seemed ingenious (if not expensive) when it was described to me in detail!

I only have one UMC adapter cable - the one that's included. Looking at the photos I can only see two variations (3 with a visible metal ring that matches and the other 5 with no visible metal ring). So now I have no way to confirm what I was told (and took on faith, coming from a Tesla employee).

I was also told that the giant 400V supply under the front hood for the HVAC was instead responsible for running the 12V accessories. Much later I learned that the 12V supply is elsewhere and the 400V comes directly from the ESS and is delivered to the HVAC system without any significant 12V involvement.

Does anyone have all of the UMC adapter cables? (actually, I seem to recall that a few members here got the full collection)

[Doug_G]

I was thinking of taking my NEMA 14-30 UMC adapter and replacing the plug with a J1772. That would be less expensive, and actually available, compared to the Tesla adapter. It would be limited to 30A but that's what most of the chargers are anyway.

[TEG]

I was thinking of taking my NEMA 14-30 UMC adapter and replacing the plug with a J1772. That would be less expensive, and actually available, compared to the Tesla adapter. It would be limited to 30A but that's what most of the chargers are anyway.

Sounds like a good plan.
But you would ignore the pilot signal from the J-plug.
There are some 15A/16A J-plugs out there - so make sure to turn down the charge rate from the VDS first...

https://www.homecharging.spx.com/volt/pdf/GM10-463.pdf

Output Voltage 208 / 240V – 15A, single phase
• Input / Output Power 3.3kW

Evr-Green Home Charging Station, Level 2, 3.8kW output, EVB22-3PM : Home Charging Station (http://store.leviton.com/Evr-Green-Charging-Station-output-EVB22-3PM/dp/B004G6ZSZG)

Level 2 - 160 Home Charging Station Cat. Nos. EVB22, EVB26 - 16A @ 120/240VAC, 60Hz

[W.Petefish]

Doug_G, take apart the 14-30 end. Look for any resistors and/or diodes. Then take apart the california end and look for the same. Pictures would be nice.

From what I saw on the 14-50 end, the purple/blue wire (I don't remember the color exactly) was connected to the green ground.

[scott451]

Does anyone know how the current revision of the UMC detects the current available from the adaptor?

I suspect that it has to do with a resistor, but what else?
Martin's RFMC design used a diode and a 90C thermal switch. My guess is that the UMC works the same way. The other possiblity is that is uses a dallas one-wire eeprom on the unused neutral pin of the california connector.

[W.Petefish]

Martin's RFMC design used a diode and a 90C thermal switch. My guess is that the UMC works the same way. The other possiblity is that is uses a dallas one-wire eeprom on the unused neutral pin of the california connector.

The neutral pin you mention is used for the blue/purple wire. Which on the UMC's 14-50 is grounded through a fuse.

[Doug_G]

Doug_G, take apart the 14-30 end. Look for any resistors and/or diodes. Then take apart the california end and look for the same. Pictures would be nice.

From what I saw on the 14-50 end, the purple/blue wire (I don't remember the color exactly) was connected to the green ground.

I haven't taken it apart, but the California connector has four connections. The outer shell connects to the NEMA 14-30 ground pin, as expected. Two of the slots connect to the hot pins on the NEMA. The third slot is, as far as I can tell, open circuit. Presumably this pin is used for signalling. I'd have to open the California connector to find out.

[samcarney]

Martin's RFMC design used a diode and a 90C thermal switch. My guess is that the UMC works the same way. The other possiblity is that is uses a dallas one-wire eeprom on the unused neutral pin of the california connector.

I purchased an RFMC before the Tesla UMC was available. It arrived with a 5-15, 10-30, and 14-50 adapter. I immediately disassembled the 14-50 and 10-30 adapters to reverse engineer.
I found that on the "California" side all 4 wires (black, red, white, green) are all connected to pins on the the California plug.
On the opposite end the black and red wires are connected to the pins on the 14-50 and 10-30 plug that conduct the charging current.
The green wire is connected to the ground pin on the plug.
The white wire is not connected to a pin on the plug.
There is a diode that is connected between the green and white wires. The orientation of the diode determines whether it is a 30 or 50 amp plug.
If the RFMC does not detect a diode (open circuit) it defaults to 16 amps.
There is also a thermal switch connected between the white and green wires. If the plug gets hot due to a bad connection etc. the switch closes and stops the charging.

So it appears that a diode connected between green and white ( I don't have my notes with me and can't remember if the positive side of the diode goes to the white wire to indicate 50 amps or 30 amps or vice versa).
An open circuit between green and white indicates 20 amps.
A short circuit between green and white indicates trouble and stops charging.

There was a posting by Martin Eberhard with a description of operation and schematics for the RFMC which helped immensely. (sorry don't have the link available either).

With this information I was able to gather the parts to build and test adapters that will allow me to plug into most receptacles in North America.
In addition to the 5-20, 10-30, and 14-50 that came with the RFMC I also have;
L5-30, 6-15, 6-20, 6-30, 6-50, 10-50, and L14-30 adapters. I built a 22 foot long 4-conductor #6 guage wire extension cord for those out of the way receptacles. I also have a cable with a "California" end on 1 side and stripped wires on the other side(pigtail) that can be connected directly to a fuse or breaker panel.

I am fairly confident I can find a compatible "plug" on road trips, I just can't take a lot of luggage

[TEG]

Yep... But does the UMC work in the same/similar way to the RFMC?

[Doug_G]

I used a DVM in "diode mode" and didn't see anything. It's possible that I simply didn't make contact, as it's hard to reach inside the slot, but I tried several times. I'll check again, but it's quite possible that the UMC works differently. As Scott said, they might use one of those "one wire" gadgets from Dallas.

[S-2000 Roadster]

I purchased an RFMC before the Tesla UMC was available. It arrived with a 5-15, 10-30, and 14-50 adapter. I immediately disassembled the 14-50 and 10-30 adapters to reverse engineer.
I found that on the "California" side all 4 wires (black, red, white, green) are all connected to pins on the the California plug.
On the opposite end the black and red wires are connected to the pins on the 14-50 and 10-30 plug that conduct the charging current.
The green wire is connected to the ground pin on the plug.
The white wire is not connected to a pin on the plug.
There is a diode that is connected between the green and white wires. The orientation of the diode determines whether it is a 30 or 50 amp plug.
If the RFMC does not detect a diode (open circuit) it defaults to 16 amps.
There is also a thermal switch connected between the white and green wires. If the plug gets hot due to a bad connection etc. the switch closes and stops the charging.

So it appears that a diode connected between green and white ( I don't have my notes with me and can't remember if the positive side of the diode goes to the white wire to indicate 50 amps or 30 amps or vice versa).
An open circuit between green and white indicates 20 amps.
A short circuit between green and white indicates trouble and stops charging.
Thanks for the report, Sam. Even if the UMC is totally different, this is still very interesting for RFMC repairs or upgrades.

[W.Petefish]

I opened up my 14-50 adaptor end. (gotta love the 8MP cellphone camera)

2477
2478

[suxxer]

here was a posting by Martin Eberhard with a description of operation and schematics for the RFMC which helped immensely. (sorry don't have the link available either).

Well, here's the document from Martin...

2479248024812482248324842485

Or the PDF:
2486

[W.Petefish]

Yep... But does the UMC work in the same/similar way to the RFMC?

From what I'm seeing, no.

[TEG]

I opened up my 14-50 adaptor end.
I can't tell from the pictures how the UMC knows which plug is attached.
Anyone see somthing?
(I wonder where the maroon/purple wires go.)

[samcarney]

I opened up my 14-50 adaptor end. (gotta love the 8MP cellphone camera)

2477
2478It appears that the maroon/purple wires are connecting the green wire to the white. (does the maroon connect to the white? It is hidden in the photo) (could there be a diode or other device between the maroon and white wires?)
If there is nothing but another wire connecting the green and white wires then it does not work the same as an RFMC. If the RFMC senses a short circuit between green (ground) and white (pilot signal) it shuts down.

[W.Petefish]

The maroon (purple actually) wire only connects to the ground on the california connector.

You can see the dead end of the wire on the 14-50 plug end if you zoom in.

[jaanton]

You will need to take apart something other than the 14-50 connector. I'll bet that grounding the neutral on that connector means full 40 amps.
In the U.S. the only amp levels the UMC will produce are 40, 24, 16, and 12. That could be done with a simple diode scheme. It would make sense that the 14-50, the most common connector, is also the simplest. I have a 6-20 connector for the UMC, I've done some simple signal tests and I'm certain there is a diode in there.

[suxxer]

Yep... But does the UMC work in the same/similar way to the RFMC?

Why shouldn't it? Martin could be the father of both of them.

You could measure the pins of the connectors to find out if there are any diodes involved. Fast & easy way to tell....

[dpeilow]

I was told that the official line on the RFMC was that it was felt to be dangerous. That could be sour grapes of course but it wouldn't suggest that ME was the father of both.

[S-2000 Roadster]

I was told that the official line on the RFMC was that it was felt to be dangerous. That could be sour grapes of course but it wouldn't suggest that ME was the father of both.
Who's to say that the official line did not originate with M.E.? He could have discovered any dangerous tendencies in hindsight and redesigned the UMC to correct them. So far, I get no clear evidence one way or the other, because either possibility seems plausible.

[suxxer]

I remember TM had problems with the "new" UMC when introduced to Europe. They said it had something to do with the bi-metal stripe inside the connector. The UMC would shout down if the connector got too hot in sunny areas. That might be evidence, that the UMC-Design can't be too far away from M.E.'s (since he's using bi-metal inside the connector as high temperature shut down too)

[donauker]

I was told that the official line on the RFMC was that it was felt to be dangerous. That could be sour grapes of course but it wouldn't suggest that ME was the father of both. The RFMC does not have a GFI or contactor so the connector pins are hot whenever it is plugged in.

[doug]

Who's to say that the official line did not originate with M.E.? He could have discovered any dangerous tendencies in hindsight and redesigned the UMC to correct them.
Highly doubtful.

[vfx]

... M.E.? He could have discovered any dangerous tendencies in hindsight and redesigned the UMC to correct them....

Martin was fixing RFMCs that he did not even make. He personally knows many of the owners of the dozen or so he crafted himself so I would be shocked (!) if he left them hanging with less than a perfect product.

[SByer]

The RFMC does not have a GFI or contactor so the connector pins are hot whenever it is plugged in.

Yeah, I've always sort of assumed this, so I typically disconnect the RFMC at the CA connector and hang that up on the wall. Also eliminates a little bit of vampire power draw. The CA connector seems to be built to take a lot of connect/disconnect cycles (so I also just leave the 10-30 side plugged in to the wall, since it's clearly not meant to take it).

[dpeilow]

Who's to say that the official line did not originate with M.E.? He could have discovered any dangerous tendencies in hindsight and redesigned the UMC to correct them. So far, I get no clear evidence one way or the other, because either possibility seems plausible.

Because he'd left the company before either product came out.

[TEG]

I recall that there was heavy demand for something like the RFMC / UMC as people were finding that the MC240 just wasn't optimal/flexible for a lot of possible charging situations. When the RFMC came out, people were excited and wanted them, but it wasn't long before the UMC came out and somewhat obsoleted the RFMC.
Perhaps the early RFMC availability forced TM to get the UMC out sooner rather than later. I don't know if the two products were independent efforts or share some similar DNA.

I am hesitant to say that the UMC obsoleted the RFMC, but it did have some benefits including:
* Direct from the car company, not a 3rd party.
* Smaller, more integrated looking design.
Are there technical details of UMC that make it a better choice than RFMC, or is it primarily aesthetics, availability, and "factory approved?"

[mpt]

Great timing; I sat down yesterday and thought, I wonder if I can make my own UMC heads? Problem is the twist lock connector that TM uses has been adapted with the metal ring. The UMC side has a metal clip protruding in in such a way to prevent standard twist lock connectors from Home Depot from being inserted without first being heavily modified physically. And that's before any electrical alterations to signal the current. Cunning stunt eh?

[suxxer]

Ok - let's stop guessing around and measure out those adapters.

We have four possible connection points iside the california twist lock connector: red ring, black ring, green ring, white ring (see pics...).

I have two of them: one 10A and one 32A adapter. (I have a SWISS Version of the UMC, so default is 10A).

The 10A adapter:
It is build according to M.E.'s 15A schematic (http://www.teslamotorsclub.com/attachment.php?attachmentid=2481&d=1314961464). Look at the following image:
Only phase (red ring, brown cable), neutral (black ring, blue cable) and ground (green ring, yellow-green cable) are connected. The third pin (white ring) is empty. The purple pilot cable is missing too.
I guess the umc defaults to 10A if the purple pilot cable is left open.
http://www.teslamotorsclub.com/attachment.php?attachmentid=2489&d=1315139277
2489

The 32A adapter:
Phase and neutral (black and red cable) are on red ring / black ring and ground is on green ring with a green-white cable. We have the purple pilot cable shorted together with the ground wire on the white ring. This adapter is not built according to M.E.'s RFMC specs.
http://www.teslamotorsclub.com/attachment.php?attachmentid=2490&d=1315139559
2490
Furthermore the purple wire is left open on the CEE32-Connector side. So it's just a shortage between ground and white ring.
http://www.teslamotorsclub.com/attachment.php?attachmentid=2499&d=1315249507
2499


The 32A is the "biggest" adapter for the SWISS UMC. If we look at the images from W.Petefish (http://www.teslamotorsclub.com/attachment.php?attachmentid=2478&d=1314941461) we can see that his 50A adapter (which is the "biggest" adaptor for US UMC's) has the same wiring: purple is shorted together with ground.

So those two cases are very clear now: the "smallest" adapter has an open purple pilot wire while the biggest adapter has the purple pilot wire shorted together with the ground wire.

I don't have other adapters at hand. Could anyone else measure out the remaining adapters(30A for US UMC's)?

[doug]

I recall that there was heavy demand for something like the RFMC / UMC as people were finding that the MC240 just wasn't optimal/flexible for a lot of possible charging situations. When the RFMC came out, people were excited and wanted them, but it wasn't long before the UMC came out and somewhat obsoleted the RFMC.
Perhaps the early RFMC availability forced TM to get the UMC out sooner rather than later. I don't know if the two products were independent efforts or share some similar DNA.

As an additional historical note, something like the RFMC/UMC is what was originally promised as the travel connector that was to come with the Roadster. By the end of the Sig100 series only the MC120 (now called the "Spare Connector") was available. Regulatory hurdles seemed to delay Tesla being able to offer even the MC240 for quite a while. Martin had been forced from the company since before even the first production Roadster had been shipped.

[W.Petefish]

Great timing; I sat down yesterday and though, I wonder if I can make my own UMC heads? Problem is the twist lock connector that TM uses has been adapted with the metal ring. The UMC side has a metal clip protruding in in such a way to prevent standard twist lock connectors from Home Depot from being inserted without first being heavily modified physically. And that's before any electrical alterations to signal the current. Cunning stunt eh?

As stated in a previous post, the twist lock connector is a standard california connector available from any electrical supply house.

Anyone have a 30A US UMC end that they would be willing to pull apart and look into?

[TEG]

As an additional historical note, something like the RFMC/UMC is what was originally promised as the travel connector that was to come with the Roadster. By the end of the Sig100 series only the MC120 (now called the "Spare Connector") was available. Regulatory hurdles seemed to delay Tesla being able to offer even the MC240 for quite a while. Martin had been forced from the company since before even the first production Roadster had been shipped.

Yeah, I think Martin mentioned doing the RFMC on his own separately from the group that did the UMC. So they had a similar goal, but different engineering.
Perhaps he was glad that the UMC came out so he didn't have to keep fielding requests for more RFMC support.

[jaanton]

Putting a meter onto my NEMA 6-20 adaptor shows a diode, small signal silicon, cathode connected to ground and anode connected to what would be the neutral.
Hope Tesla doesn't mind a little reverse engineering.

[W.Petefish]

Can you open it to confirm?

[suxxer]

Putting a meter onto my NEMA 6-20 adaptor shows a diode, small signal silicon, cathode connected to ground and anode connected to what would be the neutral.

Yes, please open it (it's only two screws..).

[TEG]

Recapping also that the Swiss UMC seems to have different current settings from the USA version.
Swiss: 10A through 32A
USA: 12A through 40A

(Correct?)

I bet the different UMCs use the same "keying" system to detect different plug types, but have different current levels depending on which version of UMC.
For instance, if you put in a Swiss end labeled "10A" into a USA UMC would it then tell the Roadster that 12A was available?

Edit: the 10A vs 12A may not be the best example as this is the "default" charge rate. Comparing how 24/32/40A plugs work is probably more relevant.

[doug]

Recapping also that the Swiss UMC seems to have different current settings from the USA version.
Swiss: 10A through 32A
USA: 12A through 40A

(Correct?)

I bet the different UMCs use the same "keying" system to detect different plug types, but have different current levels depending on which version of UMC.
For instance, if you put in a Swiss end labeled "10A" into a USA UMC would it then tell the Roadster that 12A was available?
Don't think so. I believe that minimum current is what the Roadster draws when the pilot signal is connected to ground. Swiss Roadsters default to 10 Amps since that's what is set in their firmware.

[suxxer]

I believe that minimum current is what the Roadster draws when the pilot signal is connected to ground. Swiss Roadsters default to 10 Amps since that's what is set in their firmware.

Stop believing and start knowing ;)

Tonight I pulled apart the 10A california twist lock (the one with no purple signal wires) and plugged in some lab cables to play with. I installed one on ground and one on the white rind (where the pilot wire comes in). The I tried various situations together with the UMC (see image).

http://www.teslamotorsclub.com/attachment.php?attachmentid=2503&d=1315287924
2503

So here are the results:
Wires open = 10A (this is the same way the stock 10A is wired)
Wires shorted = 32 A (this is the same way the 32A connector is wired)
Diode (Cathode on white ring) = 13A (INTERESTING!!!)
Diode (Cathode on ground) = 10A (Interesting too!!)

So what's the conclusion? Right, the Firmware of the UMC is modified to the needs of the local infrastructure (therefore the minimal current is 10A). I guess each setup (wires open, wires closed, diode one way or another) can be programmed into the UMC to generate the appropriate pilot signal.

To answer TEG's question: The roadster draws what the UMC tells him. The pilot signal is generated according to the charging standards - I don't think Tesla played with that. The UMC senses which setup is wired inside the twist lock connector and the generates the appropriate pilot signal to signal the Roadster how much current to draw. I think TM just adjusts the minimal current the Roadster draws when any pilot is absent. If a UMC is present a pilot is always generated.

Another view of the diode test:
http://www.teslamotorsclub.com/attachment.php?attachmentid=2502&d=1315287920
2502

[TEG]

I believe that minimum current is what the Roadster draws when the pilot signal is connected to ground. Swiss Roadsters default to 10 Amps since that's what is set in their firmware.

OK, so if you brought a USA spec Roadster to Switzerland, and plugged it into a Swiss UMC w/ Swiss "10A" adapter (as labeled) it would still charge at 12A...(?)
I wonder if the gray market imports had their firmware changed to charge at the lower Swiss "default" rate?

How about at the other end - if you brought a Swiss UMC to the USA, and plugged it a US Roadster with US 40AMP plug could it still charge at 40amps?


Perhaps this is all just academic, but I do think (at least from a labeling standpoint at a minimum), that they intended all the parts (plug, UMC, Roadster) to be a "matched set" for a specific region.

Worth pondering since we have both US and European owners examining their UMCs to figure out how they work. If there are actual differences in the UMC between USA and Europe it would be good to know.

[suxxer]

if you brought a Swiss UMC to the USA,

I think it's all in the firmware of the UMC. Send me an US UMC and we can proof it...

[TEG]

Stop believing and start knowing ;)
Love that quote! :love:


... (see image)...
For some reason, I cannot view your images. They aren't showing up inline.


...So what's the conclusion? Right, the Firmware of the UMC is modified to the needs of the local infrastructure (therefore the minimal current is 10A). I guess each setup (wires open, wires closed, diode one way or another) can be programmed into the UMC to generate the appropriate pilot signal...
Yes, I thought/assumed it might be like that.
So if you brought a Swiss UMC to USA, it may cause a USA Roadster to pull different current through it even if you used the USA adapter plugs.
The labels on the plugs (showing ampacity) assume that the UMC (and possibly Roadster) are all from the same region.

I knew from the RFMC scheme that there were limited possible types of plugs it could sense.
Assuming that the UMC did something similar, I thought they may want to reuse the same signal types for different plugs and current levels in other regions.

[dpeilow]

10A is the lowest Swiss outlet, 13A is the UK setting. What do other EU Roadsters default to - in countries where normal outlets are 16A?

[suxxer]

What do other EU Roadsters default to - in countries where normal outlets are 16A?

In Germany the Roadsters were set to pull 16A if no pilot was available. But TM Germany had a couple of issues with that (people blowing fuses of installations that weren't supposed to deliver 16A for 8+ hours etc.). So now the new Firmware Update sets German Roadsters to 13A (if I remeber correctly).

What I don't quite understand is why there is no 16A UMC Adapter available in Switzerland. Ok, the Firmware of the UMC has no option for 16A, but they could add the setup "cathode on ground" to the UMC's firmware to allow a 16A plug. But I think they designed the UMC for Regions, not for signle Countries.

[mpt]

As stated in a previous post, the twist lock connector is a standard california connector available from any electrical supply house.

Indeed, not available in home depot though, it's a specialist, non-standard part that has to been ordered from a bespoke supplier. And, I wasn't able to see a california one that had the metal plate inside the recess. I do believe that's just there for show or creating a firmer grip as the UMC female has a nice spring pressing against the metal plate.

[mpt]

I'm going to pick up the set; quick & simple need right now. I'll wager that the diode in the set will be a zener diode similar to the roadster end to crow-bar down the pilot to signal the EVSE that we're ready and/or need extra ventilation... though I don't know of anyone that implements that.

[W.Petefish]

Indeed, not available in home depot though, it's a specialist, non-standard part that has to been ordered from a bespoke supplier. And, I wasn't able to see a california one that had the metal plate inside the recess. I do believe that's just there for show or creating a firmer grip as the UMC female has a nice spring pressing against the metal plate.

You have to special order them from home depot.

[S-2000 Roadster]

I wasn't able to see a california one that had the metal plate inside the recess. I do believe that's just there for show or creating a firmer grip as the UMC female has a nice spring pressing against the metal plate.
I have a strong suspicion that it is not just for show and not just mechanical. I imagine that it is made of a conductive material with a conductive contact for a reason.

[donauker]

I have a strong suspicion that it is not just for show and not just mechanical. I imagine that it is made of a conductive material with a conductive contact for a reason.The metal plate and side contact are the ground connection (marked with green). The three twist lock contacts are black, red and white for L1, L2 and neutral.

[S-2000 Roadster]

The metal plate and side contact are the ground connection (marked with green). The three twist lock contacts are black, red and white for L1, L2 and neutral.
That makes sense. So, the metal plate is a safety ground, but I suppose it doesn't carry charging current. I assume the charging current flows between L1 and L2. Is Neutral separate from ground?

[TEG]

...Is Neutral separate from ground?
That may depend if you are using a 3 prong or 4 prong plug end... (?)

[S-2000 Roadster]

That may depend if you are using a 3 prong or 4 prong plug end... (?)
So far, the UMC adapters that I have inspected do not use the 4th prong (or is it technically the 3rd prong that they're not using?).

[TEG]

NEMA 14-50:
http://www.mobilebiopsy.com/imgA.gif

Since most all 14-50s you may encounter (except for some wacky cable contraptions you might find at an RV park) will have 220/240V across L1 and L2 then you don't need the Neutral.
If you had an appliance which plugs into 14-50 that wants 110/120V it could use Neutral plus one of the legs, but the Roadster charging doesn't need that lower voltage, so just using L1 & L2 is enough.
The ground, I assume, is for safety. So for the 4 prong plugs I gather they could just leave neutral disconnected. For 3 prong plugs, there isn't even a neutral to even think of hooking up... (?)

[donauker]

As a 240 volt electrical connector the green ground and neutral wires must be passed through separately per NEMA code. (They are bonded at the entrance electrical panel but isolated at subsequent panels and wiring) The green wire will not carry any current other then leakage up to any amount allowed by the GFI device but without it the car will be ungrounded and I would certainly expect a charge fault.

In the case of the Tesla as well as any other pure 240 volt device the white neutral is not needed. The RFMC and the UMC use this otherwise unneeded white contact to provide allowable current draw signals. It does this by not connecting the white California connector contact to the neutral electrical connector contact but rather tying it to the green ground connect using various options.

The RFMC adapters tie the white directly to the green to indicate a plug overheat condition by using a thermal fuse. They use a regular non-zener diode in one orientation for 24 amps and in the other orientation for 40 amps. If the white connection is unconnected it only allows 16 amps.

Since I don't have a UMC I am uncertain on their configuration but it would appear from other posts that they are not using the shorted position as a thermal condition but rather as another current level option.

[TEG]

Thanks, Don!
Nicely said.

[suxxer]

I assume the charging current flows between L1 and L2. Is Neutral separate from ground?

Just take a look at the pics. The charging current runs between HOT1 and HOT2 (from TEG's image above). Neutral is used as Pilot and Ground is used as Ground.

[mpt]

Unfortunately, Tesla are out-of-stock of the adapters that I need. So, I considered ordering some California Connectors but they're really expensive!

I feel a minor bodge coming along involving some 14-50 trailing sockets, off-the-shelf 5-15, 6-15 & 10-30 cords and careful use of the VDS current limit.

[donauker]

I built all my own adapters for the RFMC about two years ago. This was the place with the best prices for the California connectors that I found at that time. At $37.95 it was about a third of some of the prices I found.

http://www.bestmaterials.com/detail.aspx?ID=16633

[mpt]

Thanks... The game is afoot...

[TEG]

By the way, bit of trivia:
appliance info (http://www.myrv.us/electric/Pg/appliance.htm)

...In the past the body of some ovens, stoves and clothes dryers were grounded through their neutral wires, as a measure to conserve copper during the Second World War. This is no longer permitted on NEW installation's the NEC code was changed in 1996...

[mpt]

That's funny; war seems to be the only catalyst for change!

[mpt]

This weekend I made up some cables (would have preferred the Tesla ones but, still). I made 5-15/20, 6-15/20, 10-30 & duplex 5-15 to go with the existing 14-50/30 (a trick donauker taught me).

I tried the diode and found:

wht-green open: 12,15A
Diode (cathode to wht): 16A
Diode (cathode to grn): 24A
wht-green closed: 40A

I used this in my adapters to limit max current though I did see that voltage doesn't play a part and that the UMC was happy to set up 120V@40A. I tried a couple of zener diodes, 5.1V and 12V but they made no difference.

I also found out why the outlet on the front of my house doesn't work; they've transposed hot/ground on the 120v outlet; car reports the error. Same is true for the UMC, at 120v, the grounded wire has to go to YYY. Nice.

[TEG]

...duplex 5-15 to go with the existing 14-50/30 (a trick donauker taught me)...

Are you talking about the "find the two phases to get 240V from 2 120V outlets" trick?

Like one of these? :
220 240 Volts from 110 120 Volt Outlets - Catalog & Pricing (http://www.quick220.com/220_catalog.htm)

If so, you ought to have safety relays so that the 2nd plug can't be energized while you are fishing around for an outlet with the other phase.
(Possible safety hazard if done wrong.)

[TEG]

I also found out why the outlet on the front of my house doesn't work; they've transposed hot/ground on the 120v outlet; car reports the error. Same is true for the UMC, at 120v, the grounded wire has to go to YYY. Nice.

Do you have one of those simple outlet testers that can find mis-wired outlets easily?
http://ecx.images-amazon.com/images/I/41ruIGVdvcL._SL500_AA300_.jpg

[mpt]

Yeah, everyone is split-phase these days; I like the tester; "Hot Ground Reverse" - Holy crap!

[W.Petefish]

What diode did you use?

[S-2000 Roadster]

...and does the diode need to be a Zener?

For that matter, aren't Zener diodes marked with the cathode and anode in reverse? Typically, a Zener allows current to flow in both directions, but the reverse breakdown voltage is the one that's used for effect, while the normal forward voltage is not utilized.

I would suspect that a simple diode is what is needed in the charging cables, and you may just be lucky that a Zener works at all. i.e. a really low voltage Zener might confuse the charger because it allows current in both directions. Anyway, Don already mentioned that some of the chargers use a regular non-zener diode.

[mpt]

1N4001 but any cheap silicon diode would work. The zener diode that I had didn't operate any differently.

[Doug_G]

For that matter, aren't Zener diodes marked with the cathode and anode in reverse?

No. The polarity mark is always on the cathode, regardless of the diode type.

[suxxer]

wht-green open: 12,15A
Diode (cathode to wht): 16A
Diode (cathode to grn): 24A
wht-green closed: 40A

Great. Now we have the SWISS and US specs. I'm pretty sure that TM could adjust the firmware of the UMC to set the charging current for each case.

[mpt]

What I'd like to know is, do the Tesla adapters contain additional configurations to enable 20A and 30A operation?

[suxxer]

What I'd like to know is, do the Tesla adapters contain additional configurations to enable 20A and 30A operation?


According to the Homepage, the UMC should be able to accept:
15A
20A
30A
50A
(again 4 choices as the UMC accepts 4 modes (open, short, diode up, diode down))

http://webarchive.teslamotors.com/display_data/universal_chargetimes.gif

@MPT: You measured out a US-UMC, right?

[mpt]

But, when connected to a L14-30 for example, does the UMC present 24A to the car (one of its defaults) or the full 30A?

The car understands the 30A pilot signal; I was charging last night from a Chargepoint station at exactly 30A.

[Slackjaw]

But, when connected to a L14-30 for example, does the UMC present 24A to the car (one of its defaults) or the full 30A?

Not sure if this answers your question but I have the L14-30 adapter and my car only drew 24A from a line which could almost certainly have supplied more, when I was between laps at the Raceway Park road track (over here (http://www.teslamotorsclub.com/showthread.php/5877-...-at-Raceway-Park-(Englishtown-NJ)?p=75564&viewfull=1#post75564)).

[donauker]

According to the Homepage, the UMC should be able to accept:
15A
20A
30A
50A
(again 4 choices as the UMC accepts 4 modes (open, short, diode up, diode down))


The NEC code only allows usage of 80% of breaker value. So the results are as expected.

15A = 12A
20A = 16A
30A = 24A
50A = 40A

[mpt]

...my car only drew 24A...

Yep, that's good to know.

@donauker: Indeed, but, a 5-20 or 6-20, etc. might be found on circuit with a few other outlets and a 30A breaker no? Could be good to get at that extra few amps.

[donauker]

Yep, that's good to know.

@donauker: Indeed, but, a 5-20 or 6-20, etc. might be found on circuit with a few other outlets and a 30A breaker no? Could be good to get at that extra few amps.

Only in violation to the code. All wire and receptacles must be rated at or greater then the breaker. Under an overload situation the 20 amp outlet could meltdown/burn without throwing the breaker.

Also a 20 amp or greater outlet must be wired with only one receptacle per breaker.

[mpt]

Oh, that's interesting; I must check the breakers wired in my garage; I was able to pull 40A from a 5-20... For only 30 seconds whilst I tested to see if it was possible mind you. No breakers tripped and the wiring remained cool. On each side of the garage I have a string of 5-20's daisy chained through a GFCI outlet, sounds like not to code.

[donauker]

Oh, that's interesting; I must check the breakers wired in my garage; I was able to pull 40A from a 5-20... For only 30 seconds whilst I tested to see if it was possible mind you. No breakers tripped and the wiring remained cool. On each side of the garage I have a string of 5-20's daisy chained through a GFCI outlet, sounds like not to code.

Actually I got that part wrong. Receptacles rated higher then circuit are not a problem, so multiple 5-20s should be OK. The somewhat confusing one is the 5-15 which can be used for 20 amp circuits as long as there are multiple outlets per circuit. One duplex 5-15 can handle 20 amps but each outlet would be limited to a 12 amp device.

[doug]

Annoying that these things are labeled for their peak rating instead of continuous.

[S-2000 Roadster]

Annoying that these things are labeled for their peak rating instead of continuous.
I don't think it's peak versus continuous rating so much as a built-in understanding that continuous usage should only be 80% of maximum rating. In other words, I think the ratings are consistent with reality, it's just that you should never use anything at its full rating, at least not if you want to be safe. Engineering design always allows at least 10% overhead for variances and tolerances. I don't think there's a safe way to mark everything with the exact same number just to make it easy to match like numbers with like numbers. Personally, I think it's important to be aware of planning on the 80% continuous usage out of the actual 100% rating that is marked.

[mpt]

it's just that you should never use anything at its full rating

Oh, I'd better ease back on the go pedal then, 'cause I use that at 100% rating quite a lot. Perhaps this is why my rear tires are all used up after six months.

[bonnie1194]

Oh, I'd better ease back on the go pedal then, 'cause I use that at 100% rating quite a lot. Perhaps this is why my rear tires are all used up after six months.

I made it 6 1/2 months before I had to replace. :)

[S-2000 Roadster]

Oh, I'd better ease back on the go pedal then, 'cause I use that at 100% rating quite a lot. Perhaps this is why my rear tires are all used up after six months.
Ha! What I find very interesting is that Tesla Motors hides the 100% from us in so many ways.

A "full" charge is only 4.1 V, and a "range" charge is still only 4.15 V, and both of these are shy of the actual 4.2 V capacity of the batteries that any laptop would use when charging to 100%

Even with traction control turned off, the Tesla motor does not put out full torque, otherwise the wheels would spin. Rather than rely upon driver finesse, the computer manages to hold back from 100%

But, hey, I'm used to high-performance tires wearing out quickly, it's all according to their treadwear rating.

[dpeilow]

Just to reiterate - the 80% rule is true in the US but in Europe we rate things at their continuous load capability.

[doug]

I don't think it's peak versus continuous rating so much as a built-in understanding that continuous usage should only be 80% of maximum rating.
This is just convention. It's done differently (and in my opinion more sensibly) on the other side of the pond.

[TEG]

Trying to pull 80% (40amps) off of a socket labeled 50 amps at an RV park can sometimes be hit-or-miss too.
When the sockets are exposed, outdoors, and used heavily they can start to get 'dodgy' after years of use.

Sometimes a little safety margin isn't a bad thing.

[VolkerP]

It appears to me that opinions here are biased by what everyone is used to. As pointed out with the acceleration example, there are many situations where all of us want to go to 100% spec. And overclocking hardware without additional cooling is simply exploring that safety margin in a CPU design.

"What's that? I wanted my steak well done."
"Sure darling, but I thought to go on the safe side and do it just 80%"