Built a 70A PORTABLE EVSE for under $600!

[OhioEE]

As a side note, Leviton is now selling (as of late August I believe) a 30 amp J1772 plug and cord for $126 shipped which is also an option for DIYers. This is a huge price breakthrough for what should be high quality.
http://store.leviton.com/J1772-Charg...e/b/5742800011

[fishhawk]

OP: Very nice! I rebuilt my 240V Voltec using the OpenEVSE when it started flaking out due to heat problems; pretty easy to do, and worked first time. Very happy with OpenEVSE.

BTW, I didn't think SSRs were recommended for this application (look under the "Details->The bad" section):
http://code.google.com/p/open-evse/w...merica_Upgrade

[KC2TZB]

This is great information can you show the completed schematic?
Tom

I'll Try to find it. I have it somewhere.

As a side note, Leviton is now selling (as of late August I believe) a 30 amp J1772 plug and cord for $126 shipped which is also an option for DIYers. This is a huge price breakthrough for what should be high quality.
http://store.leviton.com/J1772-Charg...e/b/5742800011

Cool! I knew that these things couldn't cost nearly as much as they sold for. I think the majority of the cost is in the J1772 connector.

OP: Very nice! I rebuilt my 240V Voltec using the OpenEVSE when it started flaking out due to heat problems; pretty easy to do, and worked first time. Very happy with OpenEVSE.

BTW, I didn't think SSRs were recommended for this application (look under the "Details->The bad" section):
http://code.google.com/p/open-evse/w...merica_Upgrade

Yep, Chris Howell has done a phenomenal job with his OpenEVSE. The footprint of the EVSE board has to be seen to be believed!
If you had a small enough contactor, you could probably fit an entire 30 amp EVSE in a 4"x4" Junction Box!

As for the SSRs, I think it's more of a personal preference over them actually being a bad choice.
Many people have experience with them in heaters, where they fail often.
However, in this application I expect them to be far more reliable than a contactor.
Plus, I can switch the current to the car using the 12V supplied by the OpenEVSE without using a control relay, which eliminates another possible point of failure. It also gave me an excuse to use heatsinks, which make the whole thing look far more bad@$$

[DCFusor]

You can export to .pdf from open or libre office, both are free, and reasonably easy to use. You have your heatsinks turned sideways FWIW, for best heat dissipation (assuming you're wall mounting this), and this error more than overwhelms any superior heatsink grease you might have used. Convection is an up/down thing, not a side to side thing - you'd need fans for that. And yes SS relays do make some heat, as they have a bit of voltage drop - much more than mechanical relays.

No, I'm not saying this isn't a great thing! My own setup is somewhat custom as well...and I put my SSR's on a big vertically oriented heatsink myself. Even with the extra couple volt drop (the biggest loss in the system), they are so much more reliable than the mechanical ones, it seems way worth it. Even with no arcing on the mech relay contacts, they do fail in some million or so closures (if they are top notch, it's not uncommon to see them fail in the mere thousands).

Note, the Volt ramps up its load very gracefully, so even mech relays *might* be reliable unless you are hot-unplugging and contacts don't make/break in the right order. Else even they will never see an arc across the contacts.

[KC2TZB]

You can export to .pdf from open or libre office, both are free, and reasonably easy to use. You have your heatsinks turned sideways FWIW, for best heat dissipation (assuming you're wall mounting this), and this error more than overwhelms any superior heatsink grease you might have used. Convection is an up/down thing, not a side to side thing - you'd need fans for that. And yes SS relays do make some heat, as they have a bit of voltage drop - much more than mechanical relays.

No, I'm not saying this isn't a great thing! My own setup is somewhat custom as well...and I put my SSR's on a big vertically oriented heatsink myself. Even with the extra couple volt drop (the biggest loss in the system), they are so much more reliable than the mechanical ones, it seems way worth it. Even with no arcing on the mech relay contacts, they do fail in some million or so closures (if they are top notch, it's not uncommon to see them fail in the mere thousands).

Note, the Volt ramps up its load very gracefully, so even mech relays *might* be reliable unless you are hot-unplugging and contacts don't make/break in the right order. Else even they will never see an arc across the contacts.

I actually am leaving this on the floor, which is why I mounted them "sideways." I also left about an inch or two between the heatsink and the back of the box, so that if I do eve decide to wall-mount, there will be enough airflow to cool the SSRs. Note that this will always be placed outdoors as I do not have a garage.

I actually measured the voltage drop across the SSRs to be 1Vrms. At 15A (which is what the Volt draws), that's a 1/15 ohm resistance (V/I=R), and the power dissipated is 15 watts (P=VI or P=(I^2)*R).
15 watts is minimal since those heatsinks were designed to dissipate about 150W from CPUs (albeit with a fan).
(This problem reminds me of high school physics... haha)

It was breezy last night when I was charging my Volt, and the heatsinks were only very slightly above the ambient temperature. So again, I'm not really worried about heat dissipation.

As for the .PDF, I do remember openoffice having that functionality... I'll go ahead and give it a shot later today.

And yes, I do expect the SSRs to be very reliable, and I always unplug the car-side first before the hot-side if I ever need to move the EVSE, so I expect this setup to last a VERY long time!

[pjwood]

Bumping, because this really was a good DIY example and...

As a side note, Leviton is now selling (as of late August I believe) a 30 amp J1772 plug and cord for $126 shipped which is also an option for DIYers. This is a huge price breakthrough for what should be high quality.
http://store.leviton.com/J1772-Charg...e/b/5742800011

...if money is paramount, I see no reason OpenEVSE can't get done for <$300. I count $279 as possible for 30A units, between:
-30A relay $13
-12V power supply $45
-Current Transformer (Donut) $11
-OpenEVSE, J1772 compliant, Chris Howell board $85
-Leviton 30A 8meter plug $126
-10ga 240v Dryer dongle, $15 Home Depot

$279 is your basic total, assuming you have ~20 gauge wire, a box, solder, a volt tester, and simple electrical tools. A 30A fuse is a good idea, but at least one manufacturer on GM-Volt doesn't include internal fusing (I wired one inline). I also used a $5 ground bar, though solder/tape/a really big screw cap would probably do the trick. A higher watt solder gun (~100w) is recomended, if you're going to try heating up 10 gauge wire, but in the end the job is surprisingly straight forward. I was going to include this in my own write-up (w/pics). Faith in executing this relatively simple schematic is where confidence begins:

http://open-evse.googlecode.com/file...wer_supply.pdf

There are vitues to OpenEVSE that other options do not provide, like swapping failed parts instead of wholesale replacement, or the idea that I believe uprating the relay amperage may be all that is required for future, higher charge rate EV's. Otherwise, getting the chance to build another one actually sounds fun. What I noticed about the 70A plugs was the use of 6 gauge wire, vs 10 gauge in Leviton's 30A unit. I don't know what SPX uses, but they're to be fused at 20A, and still charge a Volt in ~4hrs. The OE charge cord is something like (14ga?/120v).

Anyone care to guess the gauge of Tesla's new Supercharger? It looks like the girth of a fuel hose, and it "pumps" a coffee-spitting 90kw:http://futuristicnews.com/vast-netwo...-electric-car/

[Neromanceres]

I don't know what SPX uses, but they're to be fused at 20A, and still charge a Volt in ~4hrs. The OE charge cord is something like (14ga?/120v).

Anyone care to guess the gauge of Tesla's new Supercharger? It looks like the girth of a fuel hose, and it "pumps" a coffee-spitting 90kw:http://futuristicnews.com/vast-netwo...-electric-car/

The OE cord is using 14AWG wire. I would guess that the SPX units are also 14AWG wire as 14AWG can be used up to 16A continuous (20A circuit).

If I had to guess for the Tesla Supercharger that would probably be 90KW @ 500VDC would be 180A. For that you would need a 3/0 (000AWG) wire minimum. That would be a heavy and large cord.

[OhioEE]

PJwood, when are you going to be able to post pics of your EVSE? I am in agreement with you.