Proximity pilot in IEC 62196

[ra...@option.no]

Hi all.

I try to figure out the requirements regarding the Proximity Pilot (yes the PP not CP or the standard proximity signal).

The PP is in type 2 (mennekes) cables used to determine max current capability of the cable, e.g. to prevent 32A charging over a 16A rated cable.

This is done with a resistor within the cable between PP and PE, and checked by the car.

However I would think there is a requirement to the EVSE to cut power if PP have a resistance over 1,5 kohms, or less than 100 ohms.

I have tested this on a couple of type 2 commercial evse's, and one of them will display an error PP message if I use a 10k resistor, while the other did not mind at all (the same evse skips diode check as well).

So I have the IEC 61851-1 standard, defining the  general requirements of charging systems, but they hardly mention the proximity at all (only to the same degree as j1772).

IEC standards are very expencive to buy, so I don't want to buy the wrong chapter, so I am wondering where in the standard I can find the document containing the summary on wikipedia: https://en.wikipedia.org/wiki/IEC_62196#Signal_pins

 

Does anyone here have some info on this?

[chris1howell .]

I have not heard of any requirement to kill power above 1.5k or below 100.

Current OpenEVSE  boards have a the hardware to support PP, the plan for the firmware was to go above and below to cover the full 6 - 80a of J1772 while hitting the values for IEC.

[Lee Howard]

For the benefit of those in the USA (because we only use tethered cables
here)... in Europe public Type 2 charging is often done with "free
cable" designs. The EVSE does not have a charging cable tethered to
it. So, the driver has a cable that they plug-into the EVSE socket and
then into the vehicle. They remove this cable when they are done.
Cables come in 13A, 20A, 32A, and 70A varieties.

The cable resistances are described in IEC61851-1 Annex B, specifically
Table B.3 "Resistor coding for vehicle connectors and plugs".

This resistance is to be checked by the EVSE. I do not know if the
vehicle is capable of checking it.

The EVSE is required to monitor the current capacity of the cable and to
never solicit a current capacity (via PWM) to the vehicle greater than
the cable supports... AND if the EV draws more current than the cable
supports it is required to interrupt the current supply.

Furthermore, the cable resistances have +/- 3% tolerances, so if a cable
resistance falls outside of those tolerances the EVSE must not provide
current as the current capacity of the cable is uncertain.

Hence, the European Type 2 EVSE that supports free-cable builds must
either have an RCBO/RCD and contactor/relay pair for each cable type
supported *or* it must have an ammeter to monitor the current being
drawn by the vehicle and interrupt the current supply if the cable's
current capacity is exceeded.

In practice, however, most EVSEs that I've seen in Europe do neither of
those. They usually only have a single RCBO/RCD and contactor/relay
pair, and usually do not have an ammeter which is monitored by the
contactor/relay controller.

Fortunately, most public Type 2 EVSEs in Europe are only 32A, and
fortunately nearly everyone uses a cable that is rated at least as much
as their vehicle will draw. So, there is relatively little likelihood
of there being a real-world problem.

Thanks,

Lee.

> --
> You received this message because you are subscribed to the Google
> Groups "OpenEVSE" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to openevse+u...@googlegroups.com
> <mailto:openevse+u...@googlegroups.com>.
> For more options, visit https://groups.google.com/d/optout.

[ra...@option.no]

So it's not a requirement.

Looking at the documentation for the ABL-Sursum (the one that cuts power) they have the following error condition:

"Resistance between FE and CS too high" (for some reason their markings are weird, I guess it's because they are german, but FE = PE (Protective earth) and CS = PP.)

But I don't see a problem with to high resistance, would only mean lower currents.

Would be worse if some moist would give a low resistance to ground, allowing for higher currents than the cable is capable off?

[ra...@option.no]

Thanks for this post, I looked again at my printout of the IEC61851-1 and did not find any table b.3 (stoped at b.2), so I went back to my laptop and found the pdf, and guess what, my prinout lacked a bunch of pages, makes much more sense now. So thanks! :)

[ra...@option.no]

So just to get this straight, it's the evse that reads the PP voltage?

So a compliant evse will act as follows: I have a CP at 50% duty cycle (30A) but add a PP/PE resistance of 680 ohm the CP will change to around 35% Duty cycle?

On Thursday, August 20, 2015 at 7:49:37 PM UTC+2, Lee Howard wrote:

[Lee Howard]

On 08/20/2015 11:06 AM, ra...@option.no wrote:
> So just to get this straight, it's the evse that reads the PP voltage?

Yes, the EVSE reads the PP resistance.

> So a compliant evse will act as follows: I have a CP at 50% duty cycle
> (30A) but add a PP/PE resistance of 680 ohm the CP will change to
> around 35% Duty cycle?

Your understanding is correct, but let me correct the details...

If the EVSE is capable of 30A but if a cable with PP resistance of 680
ohms is connected, then the EVSE must not solicit more than 20A (33.33%
duty-cycle).

There cannot be any PWM present until the PP resistance is detected and
is within tolerances.

Thanks,

Lee.

[Lee Howard]

Let me also expand on this last part...

If the EVSE is capable of 30A but if a cable with a PP resistance of 680
ohms is connected, then the EVSE must not solicit more than 20A *AND*
the EVSE must interrupt the current supply if the vehicle draws more
than 20A.

It is this last part (the interruption of current supply if the EV does
not honor the CP duty-cycle) that many EVSEs do not implement properly.

Thanks,

Lee.

[ra...@option.no]

Hi and thanks again for the info.

I see in 62196 that "Vehicle connectors and plugs using the proximity contact for simultaneous proximity detection and current capability coding of the cable assembly set shall have...." But does the standard say anything about who needs to use the current capability coding? I guess all evse that has a type 2 socket, not tethered cable?

I have yet to see any stock type 2 chargers with an amp monitoring, but I am not surprised. 

It might not be a huge problem today. But when the BMW i3 owner switches a car to a Tesla and keeps his Bmw 16A cable this will be a problem.

[Lee Howard]

On 08/20/2015 11:59 AM, ra...@option.no wrote:
> I see in 62196 that "Vehicle connectors and plugs using the proximity
> contact for simultaneous proximity detection and current capability
> coding of the cable assembly set shall have...." But does the standard
> say anything about who needs to use the current capability coding? I
> guess all evse that has a type 2 socket, not tethered cable?

Part of the fun of reading specifications like this is trying to
decipher what they mean.

Normally in Europe you're now using Type 2 (a.k.a. Mennenkes) plugs for
which the cables do not usually have a proximity switch in them. (The
car locks and unlocks the connector into the EV connector, and the EVSE
locks and unlocks the connector into the EVSE connector. The EV must
generally be unlocked and disconnected first before the EVSE lock will
release.) The proximity switch is commonly found in Type 1 (a.k.a.
J1772) plugs. However, a several years ago before Europe standardized
on the Type 2 connector this statement in B.5 made more sense upon
reading it because cables would have both the proximity switch on one
side *and* current coding on the other. Now that Europe has
standardized on the Type 2 connector it's confusing.

It *seems* to say that the current coding is only required for cables
where proximity is also present. This is not the case. Use of current
coding is required for all non-tethered ("free cable") configurations.

My copy of 61851 is from 2011. Maybe it's been reworded since then to
make more sense.

So maybe you noticed, then, that the EVSE also has to control the
locking and unlocking of the connector to the socket. It should remain
locked in States B, C, and D. This is both for safety reasons and to
prevent theft. The driver controls the lock by connecting or
disconnecting the plug at the EV side.

> I have yet to see any stock type 2 chargers with an amp monitoring,
> but I am not surprised.
> It might not be a huge problem today. But when the BMW i3 owner
> switches a car to a Tesla and keeps his Bmw 16A cable this will be a
> problem.

That scenario will only be a problem *if* the EVSE does not respect the
16A current coding in the cable. So, you have to have three things
occur: 1) a cable is used that is rated less than the EV will draw, 2)
the EVSE must support a current that is greater than the rating of the
cable, and 3) the EVSE must ignore the current coding in the cable.

I agree that it's a problem, but I will be surprised if it happens
except in situations with experimental or damaged EVSEs. (So, it should
be rare.)

I've been arguing this safety point with European EVSE developers for a
while now (that you either have to have an RCBO+contactor pair for each
supported cable rating - so a 32A EVSE would need 3 pairs, *OR* you must
have the controller monitoring an ammeter to interrupt the supply if the
EV draws more than the cable rating). Unfortunately, this point is not
commonly tested by the few EVSE test equipment out there, and the EVSE
manufacturers balk at the cost and size of adding so many RCBOs and
contactors (4 pairs for a 63/70A EVSE!).

So... as long as the EVSE controller respects the current rating of the
cable, then they'll get away with it... OR until someone makes an EVSE
tester that starts checking for this safety issue.

Thanks,

Lee.

[Craig Kirkpatrick]

I've always wondered why Europe adopted the more complex system. My theory is that there must be concern of robbers stealing cables from public EVSE stations.

Best Wishes,
Craig K

--
You received this message because you are subscribed to the Google Groups "OpenEVSE" group.

To unsubscribe from this group and stop receiving emails from it, send an email to openevse+u...@googlegroups.com.

[Lee Howard]

On 08/20/2015 02:56 PM, Craig Kirkpatrick wrote:
> I've always wondered why Europe adopted the more complex system. My theory is that there must be concern of robbers stealing cables from public EVSE stations.

Yes, that too. As I understood it, there were two thoughts:

1) The common, old perception is that the EV is like an appliance and
should come with a cable and that the utility just provides the socket.
(This changes, though with rapid DC charging which are all tethered, of
course, because nobody's going to carry-around a big, expensive cable
like that.)

2) Thieves will cut and steal cables for the metals inside. This has
already happened a few times in the UK where people cut tethered EV
charging cables... apparently for the copper.

Thanks,

Lee.

[ra...@option.no]

I think this discussion is very interresting.

The reason I stumbled upon this and my initial question is that I am building an EVSE tester.

We deliver an "add-on" module to a couple of commercial EVSE's in our country to handle load balancing.

This is basically a communication module that interacts with an existing evse controller, in addition to measuring of real amp usage pr. car, relays this info to a "master" and is told to change the duty cycle depending on how many cars charging etc. This also allows us to kill power to a car if it does not behave and draws more current than allowed, so I have now learned that we make the evse more compliant :)

We needed to make an easy to use tester for the guys in the field installing this system, so I ended up designing a tester consisting of a type 2 connector with a usb connector.

The tester was designed to test really basic things (changes states and monitor the duty cycle, to see that the load balancing was working). I now work on a revision and started adding some more tests to it, mostly because I find it interresting.

So my latest revision of the hardware has the following features:

- Manipulate (positive) CP signal to set all modes.

- Ground CP signal to get error state

- Manipulate (negative) CP signal to set diode fail state.

- Monitor duty cycle and frequency

- Line voltage check for L1-L3

- Manipulate PP signal to set current codes for "cable".

- Ground and float PP 

- Monitor PP voltage

With this our software can do a lot of testing. Obvoius once beeing checking for things like line voltage when there is not supposed to be one, duty cycle in state A, diode missing behavour, etc.

For PP testing mode I was thinking to first go State B, check available duty cycle, back to A, add suitable resistance to PP to force it lower than it initially presented, go to B and check if it comply.

I have done some "bench" testing on two different commercial evse's that I have in my lab.

As I said, one of them behaves as expected on all I threw on it, including to high/low resistance on PP. While the other one doesn't care about anything. No diode, no problem. No PP, no problem, even if it has a type 2 socket, not tethered cable. They also have left out the locking mechanism for the socket, and as I understand the standard, this is mandatory? The same unit seems to "reboot" whenever the contactor releases making it unresponsive for 10-15 seconds. The scary part is that this is the EVSE I see most in public garages.

[Craig Kirkpatrick]

Very interesting projects you are working upon. 

One USA National Laboratory has an open-source project for load balancing on the scale of a single location, expandable to a whole city. You may find it interesting and borrow some inspiration from their project. 

http://gridoptics.pnnl.gov/VOLTTRON/

I intend to deploy an example of Volttron at my home since I have two OpenEVSE potentially charging two EV at the same time.  I think I can host Volttron on a Raspberry Pi2. 

If I have free time I would like to work on the firmware for OpenEVSE to comply with the IEC standard.  Chris tells me half of his sales of OpenEVSE kits go to Europe. It seems important to finish that part of the firmware. 

I'll stay in touch with you since I could use someone independently testing OpenEVSE firmware changes to see if it is compliant with standards. 

Best Wishes,

Craig K

--
You received this message because you are subscribed to the Google Groups "OpenEVSE" group.

To unsubscribe from this group and stop receiving emails from it, send an email to openevse+u...@googlegroups.com.

[Lee Howard]

On 08/20/2015 11:51 PM, ra...@option.no wrote:
> They also have left out the locking mechanism for the socket, and as I
> understand the standard, this is mandatory?

I honestly have not seen a specification or law regarding the locking
requirement.

However, in order to deter theft and for safety purposes it should be

locked in States B, C, and D.

If it's not an actual requirement, then it should be. You can imagine
the damage to contactors and relays that it will cause if people
yank-out the cable during State C. Maybe some risk of arcing, too, if
that happens.

Thanks,

Lee.

[ra...@option.no]

Did some testing this weekend. Seems like most EVSE's does not adapt the CP depending on cable coding, instead they just give you an error and no charging if you try to use an 16A cable on a 32A charge station.

This is not very nice! Hopefully Openevse will handle this better :)

[Lee Howard]

On 08/24/2015 12:53 AM, ra...@option.no wrote:
> Did some testing this weekend. Seems like most EVSE's does not adapt
> the CP depending on cable coding, instead they just give you an error
> and no charging if you try to use an 16A cable on a 32A charge station.

By "16A cable" I expect that you mean 13A cable or 20A cable... as there
is not a resistor coding for 16A.

(I admit that I have made the same error in my communications, as well.)

The EV chargers (in the car) will support 3.3kW (16A) or 6.6kW (32A).
But the cables can be coded for 13A, 20A, or 32A. So, we often overlook
13A cables completely and simply call 20A cables as "16A". However, if
the 20A cable is used with an EV that has a 6.6kW charger on a 32A EVSE,
then the EVSE should supply 20A and not 16A.

Thanks,

Lee.

[ra...@option.no]

I have not tested this myself, but the VW e-up comes with a "16A" type 2 cable (what the actuall rating is I do not know, guess it's 13 then?) I have several reports on a local EV forum that charging from one brand EVSE set to 32A will result in an ERROR on EVSE.

I have tried this at the bench with the same evse by adding a 1,5k resistor (13A as you said) and it fails, no charging at all. While the other brand I have tested adapts the CP.

So I guess the first once are just lacy and just make it fail instead. I know this is a real pain for vw owners that has this cable. And why would they need to buy a 32A rated cable when the car can't utilize it anyways.

As mentioned, this is the same EVSE that does not care about missing diode.

[ra...@option.no]

Does anyone here know what voltages to expect between PP-PE at evse side with e.g a 32A and a 13A cable?

[Lee Howard]

On 09/15/2015 03:10 AM, ra...@option.no wrote:
> Does anyone here know what voltages to expect between PP-PE at evse
> side with e.g a 32A and a 13A cable?

Assuming you're using the same circuit as with CP...

5.000V (no external resistor detected)
3.000V (1500 Ω resistor - 13A cable)
2.024V (680 Ω resistor - 20A cable)
0.902V (220 Ω resistor - 32A cable)
0.455V (100 Ω resistor - 63A cable)

Thanks,

Lee.

[Arnis Tarassu]

Hi guys. Hard to find information about Proximity Pilot.

My question. Why are there cables for sale that have 3-phases, N, PE and thin CP and PP wires?

I supposed that EVSE should read cable/plug maximum value from the vehicle plug side. But IEC standard drawing shows that

in case of removable cable, there should be resistors inside both plugs.

EVSE reads one side, EV reads another one.

If PP wire would connect EVSE controller and vehicle PP, wouldn't that make more sense?

How about both devices measuring resistance value simultaneously.

What about tethered Type2 cable. Also no PP wire used? Can't EVSE controller use that line for some extra safety?

[EV@TucsonEV]

FYI, the J1772 Spec does not call for 5 wires in the EV cord, only 4 – L, N, Ground and Pilot. There is no provision for a proximity wire.

 

Best regards,

 

Rush Dougherty

TucsonEV

1014 E King St

Tucson AZ 85719

520 240 7493

www.TucsonEV.com

 

 

--

You received this message because you are subscribed to the Google Groups "OpenEVSE" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openevse+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

 

Virus-free. www.avg.com

 

[gnuarm.del...@gmail.com]

On Thursday, October 26, 2017 at 2:10:43 AM UTC-4, Rush - E...@TucsonEV.com wrote:

FYI, the J1772 Spec does not call for 5 wires in the EV cord, only 4 – L, N, Ground and Pilot. There is no provision for a proximity wire.

 

Best regards,

 

Rush Dougherty

Every reference I've found lists five signals including the proximity wire.  Seems this signal is used to indicate the user is disconnecting the cable from the vehicle and also tells the auto the cable is connected, so don't drive off.  In other words, vital.

  Rick C.

  - Get 6 months of free supercharging
  - Tesla referral code - https://ts.la/richard11209

[gnuarm.del...@gmail.com]

I'm still feeling my way around so I may not be right.  It looks like CCS comes in two flavors, a US version with single phase power with 5 pins and three phase which requires two more pins.   CCS-1 and CCS-2.  The CCS-1 used in the US uses the J1772 connector with two added DC pins.  The CCS-2 connector is similar using an additional 2 DC pins but the other part of the connector has some of the 7 pins for signalling.  In other words, CCS-1 and CCS-2 have some commonalities, but are different connectors.

  Rick C.

  + Get 6 months of free supercharging
  + Tesla referral code - https://ts.la/richard11209

[chris1howell]

Rick C,

There are 5 pins in the J1772 connector but only 4 wires are required to run back to the station. The proximity pin 2 resistors and Microswitch are all housed within the connector. This circuit is between the connector handle and the vehicle. There is no function on the station side so a wire in the cable is not required.

In summary:

5 pins n the connector

4 wires run back (L1, L2/N, Earth, pilot)

Proximity within handle (2 Resistors and microswitch)

Chris Howell

OpenEVSE LLC

--

[gnuarm.del...@gmail.com]

On Sunday, December 23, 2018 at 3:07:46 PM UTC-5, Chris wrote:

Rick C,

There are 5 pins in the J1772 connector but only 4 wires are required to run back to the station. The proximity pin 2 resistors and Microswitch are all housed within the connector. This circuit is between the connector handle and the vehicle. There is no function on the station side so a wire in the cable is not required.

In summary:

5 pins n the connector

4 wires run back (L1, L2/N, Earth, pilot)

Proximity within handle (2 Resistors and microswitch)

Chris Howell

 

Yes, thank you.  I've been reading more and realize I was confused by a drawing I saw on the web.  It's on the Wikipedia page and I believe it comes from some document, not necessarily a standard.  Anyway, this document does not clearly indicate what is connector and what is cable so I missed that the proximity circuit is only in the handle and does not go through the cable to the EVSE unit. 

So in the European version with the detached cable, is the same type connector used on each end of the cable, one end male, one female?  Then is the proximity pin used on the EVSE end to signal the cable current capacity?

  Rick C.

  -- Get 6 months of free Supercharging  
  -- Tesla referral code - https://ts.la/richard11209

[Mike Van Emmerik]

On Tuesday, 25 December 2018 09:12:12 UTC+10, gnuarm.de...@gmail.com wrote:

So in the European version with the detached cable, is the same type connector used on each end of the cable, one end male, one female?

Sort of. The one at the EVSE end has a longer shroud (outer plastic) than the one in the vehicle. They do this so that you can't daisy chain two type 2 cords (i.e. use an "extension cord"). This is annoying for those of us with legacy J1772 vehicles, because when we use an adapter cord for non-tethered EVSEs, it doesn't work on EVSEs with tethered cords. This means that you have to use two different cords for tethered and non-tethered EVSEs, or use a special short extension cord anyway. Outside North America, Tesla destination chargers are essentially tethered type 2 EVSEs, and are quite common. So if (as I have now) you only have the cord for non-tethered type 2 EVSEs, you miss out on a lot of charging opportunities.

 

  Then is the proximity pin used on the EVSE end to signal the cable current capacity?

Yes, that's my understanding. My type 2 to type 1 adapter cable has 680Ω from PP to GND at the type 2 EVSE end (signifying 20 A, though it's advertised as a 16 A cord), and the usual 150Ω or 480Ω when the switch is pushed at the vehicle end (J1772 / type 1). There is no connection from PP at one end to PP at the other (other than through GND, of course).