UMC vs SPC - how can SPC get away without the "box"?

[smorgasbord]

The Spare Mobile Connector is just a wire and the connector:

The Universal Mobile Connector has a big box along the wire:

What technically is going on that the Spare Mobile Connector gets away without such a box?

What are the limitations of the Spare Mobile Connector? Could I feed 20 amps through it at 110volts? Could I feed 15 amps through it at 220volts?

What's driving this is that I stayed at a friend's beach house this past weekend, and they only have 110volt outlets. I was thinking about getting the Quick 220 thing to piece two 110volt out of phase outlets into a single 220volt outlet, but then I would have had to have brought my UMC, and then the total package was getting alot bigger than I had room for in the Roadster trunk. So, I paid a visit to the SLO 70 amp chargers and spent a couple hours at Starbucks doing work.

 

[stopcrazypp]

The SPC is a completely dumb cable that's only for low power charging modes (where an EVSE is not required by the electrical code).

The electrical code requires an EVSE for any EV charging connection except for 110V single phase sockets:
http://www.madkatz.com/ev/nec1999Article625.html

The Roadster seems to be the only EV that even supports 110V charging without a EVSE (the Model S doesn't seem to support this).

So the box in the UMC serves as a mobile EVSE and also has logic inside to decode the various adapters and translate it into the proper J1772 current/voltage signaling on the car end.

 

[slcasner]

What technically is going on that the Spare Mobile Connector gets away without such a box?

The small pin in the Roadster's charging connector carries the pilot signal from the EVSE to the car telling how many amps the EVSE can provide. The signal is a square wave where the duty cycle indicates the maximum available current. With the SMC, there is no square wave on that pin, just a resistor to ground or maybe it is just open. The Roadster accepts that signal to mean that the current limit is 16A.

The EVSE includes a circuit powered by the line voltage to generate the signal and also sequence through some handshake steps with the car (see https://code.google.com/p/open-evse/wiki/J1772Basics). So, the box has to include that circuit, plus its power supply, plus a GFCI big enough to handle the higher current. The wall-mounted HPC (at least the Roadster one, not sure about the Model S one) also has a big contactor (relay) that only closes to allow the line voltage through to the car connector pins after the handshake completes. The UMC and the older Roadster mobile cable don't have the contactor.

What are the limitations of the Spare Mobile Connector? Could I feed 20 amps through it at 110volts? Could I feed 15 amps through it at 220volts?

Nope and nope. You may notice that when you have the SMC connected to the car the current adjustment in the car goes only to 16A. I tried an experiment where I connected 220 volts across the line and neutral pins of my SMC (on which the GFCI has been replaced with a good-quality plug), but the car would not charge. I assume the software only allows the null pilot signal when the voltage is 110.

 

[hcsharp]

What technically is going on that the Spare Mobile Connector gets away without such a box?

It's easier to first answer what's going on in the UMC box. Your Roadster is capable of pulling 70A from power supply when you plug it in. If it did that every time you plugged it in, it would burn a lot of houses down. Or at least trip a lot of breakers. So it needs to know how many amps it is allowed to pull. The so-called pilot signal on the small pin lets it know how many amps are available. The UMC box contains a smart pilot signal generator. I say "smart" because it determines the number of amps to signal based on which adapter you are using with it. The UMC also contains a large relay to connect/disconnect all power. I won't bore you with the reasons for the large relay.

Generally your roadster will not charge without a pilot signal letting it know how many amps it can pull. There is an exception to this, and that is charging at 115v, up to 16A. The EU exception will do 13A at 230v. If the roadster senses 110v and a grounded pilot signal, it will start charging at 15A (or is it 16?). That is what the SMC provides: 115v and a grounded pilot, so the roadster starts charging at 15A.

What are the limitations of the Spare Mobile Connector? Could I feed 20 amps through it at 110volts? Could I feed 15 amps through it at 220volts?

Yes, and yes, but not legally and not without modifications. The modifications would require that you add a pilot signal generator to pull more than 16A at 120v. Or pull anything at 240v. But after the modifications you would essentially have the same thing as a UMC. The only point of this exercise would be to make yourself a low-powered UMC, and even then it would violate code if you pulled more than 16A.

What's driving this is that I stayed at a friend's beach house this past weekend, and they only have 110volt outlets. I was thinking about getting the Quick 220 thing to piece two 110volt out of phase outlets into a single 220volt outlet, but then I would have had to have brought my UMC, and then the total package was getting alot bigger than I had room for in the Roadster trunk. So, I paid a visit to the SLO 70 amp chargers and spent a couple hours at Starbucks doing work.

When I stay with friends/relatives on long trips without a dryer or stove outlet, I install a temporary 240v 50A breaker in their panel to a NEMA 14-50R. But I have a fair amount of electrical experience and know how to do this safely. It only takes me a few minutes. There are other options if I can't bring the right breakers for their panel. That discussion will wait for another day. I've even stayed at a number of Airbnb hosts this way. Of course I checked it out with them ahead of time and paid extra for the electricity and general special need.

 

[slcasner]

Henry, I don't know if you read my post before writing yours, but I note that our posts are contradictory in a few places.

I confess that I have never seen the inside of a UMC, so my statement that it does not contain a contactor was a guess. I do know that the original Roadster 30A 240V Mobile Cable with the large, rectangular box did not contain a contactor. It did contain a rather large 30A GFCI, which might be considered a relay, but the GFCI did not open to disconnect power unless there was a ground fault. Martin Eberhard's RFMC contains neither a contactor nor a GFCI. Since the UMC box is much smaller, I made the assumption that it does not contain a contactor either.

You've clarified that the SMC pilot signal's "resistor to ground" is 0 ohms, AKA a wire. OK.

Most interesting are our contradictory answers to the questions about extending the functionality of the SMC. I interpreted smorgasbord's questions to be about handling 20A/110V or 16A/220V without modifying the SMC (save perhaps replacing the GFCI plug with a 220V plug); that is, without adding a pilot signal generator that would require a box to contain it. Drawing 20A isn't allowed by the car's software so that is not possible unless the OVMS can override it. However, drawing 20A from a 20A circuit will cause the breaker to trip, so you'd really need to step up to a 30A outlet (which needs a different connector). For the second case, 16A/220V, I know that won't work without a pilot signal because I tried it.

 

[hcsharp]

Henry, I don't know if you read my post before writing yours, but I note that our posts are contradictory in a few places.

I confess that I have never seen the inside of a UMC, so my statement that it does not contain a contactor was a guess. I do know that the original Roadster 30A 240V Mobile Cable with the large, rectangular box did not contain a contactor. It did contain a rather large 30A GFCI, which might be considered a relay, but the GFCI did not open to disconnect power unless there was a ground fault. Martin Eberhard's RFMC contains neither a contactor nor a GFCI. Since the UMC box is much smaller, I made the assumption that it does not contain a contactor either.

You've clarified that the SMC pilot signal's "resistor to ground" is 0 ohms, AKA a wire. OK.

Most interesting are our contradictory answers to the questions about extending the functionality of the SMC. I interpreted smorgasbord's questions to be about handling 20A/110V or 16A/220V without modifying the SMC (save perhaps replacing the GFCI plug with a 220V plug); that is, without adding a pilot signal generator that would require a box to contain it. Drawing 20A isn't allowed by the car's software so that is not possible unless the OVMS can override it. However, drawing 20A from a 20A circuit will cause the breaker to trip, so you'd really need to step up to a 30A outlet (which needs a different connector). For the second case, 16A/220V, I know that won't work without a pilot signal because I tried it.

Actually, I don't think we contradicted on much at all. The GFCI inside the original 30A mobile charger did contain a relay but like you said, it never interrupted current without a ground fault, so its operation as a relay could almost be ignored. By contrast, the UMC has a much bigger relay that turns on and off every time you plug in.

My answer about using the SMC may have appeared to contradict yours, but I qualified it with "but not legally and not without modifications." So in the end I think we had pretty much the same answer to that. Considering that it would be a waste of time to make the modifications necessary, and there would be serious safety issues, I probably should have provided a response more like yours. I might have been influenced by a handful of Roadster owners who have actually modified their SMC to use it as a more compact mobile charger.

 

[smorgasbord]

Thanks for all the info.

My question was about using it with a Quick 220 in order to get something like 15 amps at 220 volts. As it stands, if I had to bring along my UMC and the Quick 220, I wouldn't - I need the trunk space for other things.

 

[hcsharp]

Thanks for all the info.

My question was about using it with a Quick 220 in order to get something like 15 amps at 220 volts. As it stands, if I had to bring along my UMC and the Quick 220, I wouldn't - I need the trunk space for other things.

The OpenEVSE board is very small and I wouldn't be surprised if it fit in the Quick 220 box. Just a thought. You might also be able to find one of the original Juice Boxes. They weren't the best EVSE design but they were very small (the newer ones are better, but bigger). You could have it plug into the output of the Q220 and have it programmed to 15A.

 

[slcasner]

There's no pilot wire in the SMC cable, though. You'd have to open up the Tesla connector (the car end) and hook in a new wire that goes to wherever you put the OpenEVSE board. If it was in the Q220 box at the other end of the SMC cable, that would mean you'd need to run a new wire along the length of the cable.

 

[hcsharp]

There's no pilot wire in the SMC cable, though. You'd have to open up the Tesla connector (the car end) and hook in a new wire that goes to wherever you put the OpenEVSE board. If it was in the Q220 box at the other end of the SMC cable, that would mean you'd need to run a new wire along the length of the cable.

Darn I forgot about that. Yeah, the only reasonable option is use the SMC with 115v if he doesn't want to bring his UMC. Just stay longer at the beach house!

 

[TEG]

Some vehicles, like LEAF and Volt I think, go through a test sequence before they start charging from an EVSE. This includes making sure the contractor works properly to disconnect the line current when signaled to do so. For that reason, EVSEs without a contactor (like the old RFMC) would not work to charge those vehicles. If the charging system in the car doesn't check much, you can get away with leaving off safety features on the EVSE and still have it work. Although the basic function of an EVSE is fairly simple, there are a surprising number of incompatibilities between different vehicles and different EVSEs. On Roadster, differences between 1.5 & 2.x charging systems had them behave differently with some EVSEs. In some cases there were differences in the amount of current leakage they would allow before the GFCI features would activate.

Also, just because an EVSE has a small "box" doesn't mean it lacks a contactor / relay. For instance, the Model S UMC has a little tiny one.

 

[donauker]

I do know that the original Roadster 30A 240V Mobile Cable with the large, rectangular box did not contain a contactor. It did contain a rather large 30A GFCI, which might be considered a relay, but the GFCI did not open to disconnect power unless there was a ground fault.

Some vehicles, like LEAF and Volt I think, go through a test sequence before they start charging from an EVSE. This includes making sure the contractor works properly to disconnect the line current when signaled to do so. For that reason, EVSEs without a contactor (like the old RFMC) would not work to charge those vehicles.

I can say that one of these statements is not 100% correct. I modified my original Roadster 30A 240V Mobile large box EVSE by removing the Roadster connector and cable and replacing it with a J1772 connector and cable. This unit is being used to charge a 2014 Leaf on a daily basis.

 

[qwk]

Darn I forgot about that. Yeah, the only reasonable option is use the SMC with 115v if he doesn't want to bring his UMC. Just stay longer at the beach house!

If one knows what they are doing, they can modify/make another SMC to charge without a relay/contactor. Scott did that and used it at up to 60A 240V(pic below).

 

[hcsharp]

I can say that one of these statements is not 100% correct. I modified my original Roadster 30A 240V Mobile large box EVSE by removing the Roadster connector and cable and replacing it with a J1772 connector and cable. This unit is being used to charge a 2014 Leaf on a daily basis.

I charged a Leaf once with a portable charger with no contactor/relay in it. I've also been told that not all Leafs are tolerant of this. Maybe the early years were different? Both the MS and Roadster will charge without a contactor. The Ford Focus Electric will not charge without a contactor/relay.

- - - Updated - - -

If one knows what they are doing, they can modify/make another SMC to charge without a relay/contactor. Scott did that and used it at up to 60A 240V(pic below).

This is what I meant by "modifications" to the SMC. smorgasbord could do the same with his cable. That little control box could hang out of the roadster connector eliminating the need for the extra wire in the cable. Would it be worth his time? No. Although I don't know how often he would use it. I've heard from a couple other roadster owners who've converted their SMC this way. One of them used it at 24A with a TT30 outlet on a regular basis. It's a very compact charging solution, It's just not legal or necessarily safe.

 

[TEG]

... I do know that the original Roadster 30A 240V Mobile Cable with the large, rectangular box did not contain a contactor. It did contain a rather large 30A GFCI, which might be considered a relay, but the GFCI did not open to disconnect power unless there was a ground fault..

I can say that one of these statements is not 100% correct. I modified my original Roadster 30A 240V Mobile large box EVSE by removing the Roadster connector and cable and replacing it with a J1772 connector and cable. This unit is being used to charge a 2014 Leaf on a daily basis.

It was my understanding, from things I read on the forum, that the MC240 could turn the 240V on/off using the contactor in the GFI module:

Extra wire on MC240 GFCI

My first mc240 went bad a few months after I got my roadster. The red wire connects to an expensive Solid state relay (actually it's a optically connected triac)....

IIRC. The extra red wire is a 240v control for the contactor inside the gfi. The pilot board uses this signal to turn on/off the 240v to the car.

So the GFCI itself may have only turned off the 240v in case of a ground fault, but it sounds like there was an extra control line to let the pilot board switch it on/off too.

- - - Updated - - -

I charged a Leaf once with a portable charger with no contactor/relay in it. I've also been told that not all Leafs are tolerant of this. Maybe the early years were different? Both the MS and Roadster will charge without a contactor. The Ford Focus Electric will not charge without a contactor/relay.

Yes, there have been a number of interoperability issues between different versions of different vehicles, and different types of J1772s EVSEs.
Like I said, the basic operation is fairly simple, but the safety checks they do during initial handshake can something cause unintended failures.

Here was a bit more about that:
http://priuschat.com/threads/charging-pip-using-chevy-volt-110v-charge-cord.106366/

...we confirmed that the first 2 versions of the Volt 120V charge cords do not work with the 2012 PiP. I hope to upgrade to the newest 3rd version of the Volt charge cord soon and will check that with the PiP as soon as I get it. The PiP 120V charge cord worked fine charging the Volt.
GM and Nissan fixed their interoperability problems in the 2012 model year. The 2011 LEAF cord works now with the 2012 Volt where it doesn't with the 2011. Likewise, the new 2012 LEAF charge cord now works with the 2011 Volt where the 2011 LEAF charge cord used to fail...

 

[donauker]

It was my understanding, from things I read on the forum, that the MC240 could turn the 240V on/off using the contactor in the GFI module:

Extra wire on MC240 GFCI


So the GFCI itself may have only turned off the 240v in case of a ground fault, but it sounds like there was an extra control line to let the pilot board switch it on/off too.

It has been a number of years since I personally used my MC240 but my fading memory would tend to agree with you on this. I believe there was a click from the box upon initiation of charge by the car. Just need to remember to test it out when it is no longer needed by its current user.

 

[rudholm]

Does anyone know how the UMC knows which NEMA dogbone is connected to it? I was thinking I might need to make one or two of my own since Tesla doesn't have them anymore (I kinda want a NEMA 10-30 in my kit). The UMC somehow knows when I connect a 14-50 vs a 14-30 and tells the car the correct maximum current draw, but it's not obvious how it's knows since each dogbone has the same CR6364 (which looks like a CS50 to me) on the downstream end.

Are they doing something clever, like an embedded RFID or something?

 

[gregd]

Yeah, I've been wondering that too... Time to break out the meter.

I have both the 14-50 and 10-30 dogbones, and they appear to be doing something between the ground plate on the side of the connector and the neutral slot on the connector. On my 14-50, they are tied together, while on the 10-30 there appears to be a diode, cathode to neutral, between them. I'm going to guess that various charging levels are determined either by a resistor in series with the diode, or that the diode is actually a zener, with different voltage levels.

Forward voltage is about 0.57v, if I trust my meter. I don't have a convenient / safe way to determine the reverse breakdown of the diode, if it is a zener. If there is a resistor in series on the 10-30 dogbone, it's either really small (few ohms), or not present for this configuration (zero ohms); I'm going to guess zero. You might try just a simple 1N400x diode all by itself and see what the car thinks the charging current would be. The 10-30 should limit the max to 24 amps.

For completeness, the ground plate on the side of UMC-end ties to the ground pin on both the 14-50 / 10-30 end. Neutral on the 14-50 is not connected, and is, of course, not present on the 10-30. The two hot leads are sent straight through.

Good luck!

 

[rudholm]

Yeah, I've been wondering that too... Time to break out the meter.

I have both the 14-50 and 10-30 dogbones, and they appear to be doing something between the ground plate on the side of the connector and the neutral slot on the connector. On my 14-50, they are tied together, while on the 10-30 there appears to be a diode, cathode to neutral, between them. I'm going to guess that various charging levels are determined either by a resistor in series with the diode, or that the diode is actually a zener, with different voltage levels.

Forward voltage is about 0.57v, if I trust my meter. I don't have a convenient / safe way to determine the reverse breakdown of the diode, if it is a zener. If there is a resistor in series on the 10-30 dogbone, it's either really small (few ohms), or not present for this configuration (zero ohms); I'm going to guess zero. You might try just a simple 1N400x diode all by itself and see what the car thinks the charging current would be. The 10-30 should limit the max to 24 amps.

For completeness, the ground plate on the side of UMC-end ties to the ground pin on both the 14-50 / 10-30 end. Neutral on the 14-50 is not connected, and is, of course, not present on the 10-30. The two hot leads are sent straight through.

Good luck!

Thanks for the info. I verified that on my 14-50 dogbone, the Neutral pin on the 14-50P end is not connected. That's actually handy, because it means I can remove that pin and use that one dogbone on 14-30, 14-50, and 14-60 receptacles (I just have to remember to limit current to 24A if I plug into a 14-30R). Those three connectors are the same other than the Neutral pin.

I still need to find/make a 10-30 dogbone, though.

 

[rudholm]

Hmm, on my 14-30, I see what looks like a diode, cathode on neutral, but I see ~2.3Mohms in the conducting direction (and no conduction the other direction). I expected it to look the same as your 10-30 since both are 30A dogbones. I really wish Tesla would publish more technical data.