Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

DIY Home EVSE

This site may earn commission on affiliate links.
#141
#144
I wanted to post a pic of my latest DIY 70 amp HPC. It's mounted outside and I wanted to give it a unique, decorative kind of look. When I saw this attractive cookie tin I realized it fit the bill. The stoplight-looking buttons raise and lower the amps of the pilot signal. Much easier than climbing in and out of the car and cycling through menus. Plus I'm not limited to 8 amp increments and I can change it on the fly even while it's charging. The green button is a quick cool-down button (thanks to Scott for giving me this idea). If you get home with a warm battery but don't want to charge right away (TOU metering), hit the green button and it will charge at a minimum rate long enough to cool your battery down so it won't sit soaking at a high temp waiting for off-peak rates. The car will automatically start charging at it's normal scheduled time and amperage later in the night.

IMGP0348.JPG


I made everything from scratch. OK, I bought the 100 amp contactor on ebay ($40 new surplus). What's kind of amazing about this device is the variety of countries that all the parts originated from. Here's a partial list in addition to China, Malaysia, Philippines and others :

Cookie tin: Denmark
PCB: USA
Enclosure clear top: Finland
Contactor: Sweden
Switch (red slide on connector): Costa Rica
Design and final assembly: Vermont, USA
 
#145
Very cool. Thanks for sharing. I like the user interface and features.

GSP

PS. We are living on a small world these days. I think this enables many innovations, accelerating human culture and technology.
 
#148
Robert.Boston said:
I love the repurposing of the cookie tin. There's an extra element of fun that it reads "Hans Christian Andersen", the great author of fairy tales. Shall we read a parallel in this to the rise of Tesla?
Click to expand...

Yeah Hans was part of what made this project fun. It took a bit of work. I had to reinforce the cookie tin on the inside to make it strong enough and still be fire proof for the high-voltage components inside.

GSP said:
Very cool. Thanks for sharing. I like the user interface and features.

GSP

PS. We are living on a small world these days. I think this enables many innovations, accelerating human culture and technology.
Click to expand...

Most economists agree, despite a few bumps along the way. Now if globalization would accelerate the obsolescence of fossil fuels...
 
#150
dsm363 said:
Wow. That looks awesome. You should market that.
Click to expand...

I've made a few Tesla connectors like you see in the picture and would be happy to sell them to anyone working on their own projects. I've sold a few already along with the short and sweet J-1772 adapters that I made. The newest version of that has a lock so no one can remove it from your car. I don't make any money on these components - just doing it to help people out. I also have UL listed EVSE cable rated for 80 amps if anyone needs it. Also happy to share schematics and source code.

@dsm363: If you want one just send me a cookie tin of your liking or other attractive enclosure. Maybe send me a picture of it first so I can give you a quote and determine if it's suitable. I made one on the inside of an old metal Tonka truck. Ironically it was a Texaco Tanker! But looked cool with a 50's style body.
 
#151
hcsharp said:
@dsm363: If you want one just send me a cookie tin of your liking or other attractive enclosure. Maybe send me a picture of it first so I can give you a quote and determine if it's suitable. I made one on the inside of an old metal Tonka truck. Ironically it was a Texaco Tanker! But looked cool with a 50's style body.
Click to expand...
Thanks. I'll definitely keep it in mind. Just using a UMC at the moment.
 
#152
Chris1howell said:
Any issues charging the roadster with the Open EVSE?
Click to expand...
I finally got around to testing an early Open EVSE board with my Roadster. Initial results were an error (flashing green LED) and I've fed back detailed results.

Anyone here had any issues with low pilot signal levels upsetting the Roadster (10.6V to 11.4V)?

Any clues or Roadster issues that we can give the Open EVSE folks?
 
#154
Chris1howell said:
Flashing green means "EVSE Ready - Waiting for EV"
Click to expand...
The very early Open EVSE board you sent me would not communicate with the Roadster... we moved over to the Open EVSE V1 board and had some fun because of a dry joint on the xtal (the board would run [internal clock source?] but would not charge). Once resolved it looks to be working just fine. I will update to 0.5.0 firmware and let you have some more feedback and photos :smile:

We've spent some time on this now and I think the issue with the Open EVSE V1 board is the Roadster is intolerant of pilot signal voltage errors (unlike the Leaf). The V2 board works ok at ambient temperatures but again does not meet J1772 Voch and Vocl specs (and I suspect only works because this board has high 12V supplies from the onboard DC-to-DC converter). My recommendation is to move to a rail-to-rail op amp and watch out for tolerance issues with the DC-to-DC.
 
Last edited:
#156
We've spent some time on this now and I think the issue with the Open EVSE V1 board is the Roadster is intolerant of pilot signal voltage errors (unlike the Leaf). The V2 board works ok at ambient temperatures but again does not meet J1772 Voch and Vocl specs (and I suspect only works because this board has high 12V supplies from the onboard DC-to-DC converter). My recommendation is to move to a rail-to-rail op amp and watch out for tolerance issues with the DC-to-DC.
Click to expand...

I tested a bunch of boards I have including one from the same batch as yours. All were well within the limits for Voch (nominal 12.0 Min 11.40 max 12.60) all the boards I tested fell between 11.87 and 11.92. They were all a little low for Vocl (nominal -12.0 Min -11.40 Max -12.60) averaging 11.38... Being slightly low for vocl should not matter at all because the negative portion of the pilot is killed by a diode.

J1772:2010 does not specify voltages requirements for each state but provides a table from previous J1772 versions as a reference.
All the boards I tested were at the lower range of voltage table for each state.

Roadster is intolerant of pilot signal voltage errors (unlike the Leaf).
Click to expand...
It appears the Roadster is indeed more sensitive, while OpenEVSE does not appear to be out of spec (except for vocl) the roadster seem more picky than the Leaf, Volt, i-MiEV, and PiP...

My goal for OpenEVSE is to be the best and most compliant EVSE out there period so I will be happy to work with the TESLA community to get OpenEVSE and the Roadster more compatible. I agree that a Rail to Rail Opamp and tighter regulation of the pilot voltage would be the optimal solution. I may also be able to pull up the voltages by reducing the pilot resistor so that the equivalent resistance (and voltage drop) of the current Opamp plus R1 is closer to 1k total.

I am not too far from efxjim, so maybe I can work with him to get this sorted out...
 
#157
I got my Open EVSE (Thanks Chris!) and my intention is to build a 3-phase 32A EVSE with it (pics will follow, this one will be cool!)

Anyway, I have the one with the Advanced Power Supply and I checked this schematic: http://code.google.com/p/open-evse/downloads/detail?name=Advanced_Power_Supply_Schematic.pdf&can=2&q=

That one is based on a 110/220V network where you use L1 and L2 to get 240V, however, in the EU we have 230V between L1 and N, so I'm going for L1 and N to power the ADVPWR.

There is one problem I think (not tested yet) with the CFGI test. L1 and L2 go back to TEST_L1 and TEST_L2, but in my case I would "route" L1 and N back.

The code of Open EVSE should be modified I think?

#ifdef ADVPWR
int PS1state = digitalRead(ACLINE1_PIN);
int PS2state = digitalRead(ACLINE2_PIN);

if (chargingIsOn()) { // ground check - can only test when relay closed

if ((PS1state == HIGH) && (PS2state == HIGH)) {
// bad ground

tmpevsestate = EVSE_STATE_NO_GROUND;
m_EvseState = EVSE_STATE_NO_GROUND;

chargingOff(); // open the relay
nofault = 0;
}
}
else { // stuck relay check - can test only when relay open
if ((PS1state == LOW) || (PS2state == LOW)) {
// stuck relay

tmpevsestate = EVSE_STATE_STUCK_RELAY;
m_EvseState = EVSE_STATE_STUCK_RELAY;

nofault = 0;
}
}
#endif // ADVPWR
Click to expand...

I should only check if PS1state is HIGH or LOW? That would however trip my own CFGI/Gound fault protection since the EVSE will leak some power back to the GND, correct?

I'm trying to figure this one out so I can make a nice Wiki about building you 3-phase EVSE with this project :)
 
#158
widodh said:
I got my Open EVSE (Thanks Chris!) and my intention is to build a 3-phase 32A EVSE with it (pics will follow, this one will be cool!)

Anyway, I have the one with the Advanced Power Supply and I checked this schematic: http://code.google.com/p/open-evse/downloads/detail?name=Advanced_Power_Supply_Schematic.pdf&can=2&q=

That one is based on a 110/220V network where you use L1 and L2 to get 240V, however, in the EU we have 230V between L1 and N, so I'm going for L1 and N to power the ADVPWR.

There is one problem I think (not tested yet) with the CFGI test. L1 and L2 go back to TEST_L1 and TEST_L2, but in my case I would "route" L1 and N back.

The code of Open EVSE should be modified I think?



I should only check if PS1state is HIGH or LOW? That would however trip my own CFGI/Gound fault protection since the EVSE will leak some power back to the GND, correct?

I'm trying to figure this one out so I can make a nice Wiki about building you 3-phase EVSE with this project :)
Click to expand...


Hi widodh... I am glad you recieved the OpenEVSE parts. Your Advanced PS has 44k resistors to handle the 230V so the only change would be in the firmware.
All you would need to do is replace svclvl = 1 in the ADVPWR section with svclvl = 2. That way if it detectd 230V on either test line you will get the correct pilot.

You are correct the test may trip an upstream Ground Fault circuit as each test leaks a very small current to ground.
 
#159
Chris1howell said:
Hi widodh... I am glad you recieved the OpenEVSE parts. Your Advanced PS has 44k resistors to handle the 230V so the only change would be in the firmware.
All you would need to do is replace svclvl = 1 in the ADVPWR section with svclvl = 2. That way if it detectd 230V on either test line you will get the correct pilot.

You are correct the test may trip an upstream Ground Fault circuit as each test leaks a very small current to ground.
Click to expand...
I just checked, but my code (SVN checkout) says something different (line 1112):

Code: 
#ifdef ADVPWR
char doPost()
{
  char svclvl = 0;

  pinMode(RED_LED_PIN, OUTPUT);
  pinMode(PILOT_PIN, OUTPUT);
  pinMode(CHARGING_PIN, OUTPUT);
  pinMode(VOLT_PIN, INPUT);

It's 0 there, not 1. Did you mean that code block?
 
#160
You must log in or register to reply here.

Similar threads

lencap
Help Me Confirm the my Tesla Wall Connector is Wired Correctly
Replies
26
Views
4K
Supercharging & Charging Infrastructure
1965Falcon
1965Falcon
unharmed
Australian standards for home charging an Electric Vehicle.
Replies
75
Views
3K
Australia
Nickdp
Nickdp
T
Home charging and a NEMA 14-30 adapter for mobile charger ok?
Replies
6
Views
392
Battery Discussion
ucmndd
ucmndd
R
New Owner--'23 Model 3 standard. First charge-charging door wouldn't shut.
Replies
8
Views
469
Model 3
Rob42
R
U
Charging battery properly
Replies
17
Views
397
Model S
Bullene
B
Top
Back
Blog
New
Forum list
Marketplace
Menu