High temp behavior openevse

[Danny ter Haar]

It is *hot* out here between CA and OR
yesterday was 97-100 during the day.
we stopped today at RV park to grab a few hours of charge. Learning f4on my mistake a week ago I put the openevse in the shade.
at some point i look at the display of my car what it is charging (231 volts 21 amps)
Looked at the openevse and see the display is 50C and limiting the current to 20 amps. I opened up the openevse case and put a little rock in between to lower the temp. It now shows 43C and supplies 40 amps again to my EV. I will post pictures later, but i am thinking of getting a different casing with a temp controlled fan. This is no fun ;-)

[chris1howell .]

The current temperatures for throttling may be too consecutive we may be able to increase the points by 10c without issue. We have never seen reports of instability due to heat so I think we can raise the limits a bit.

With the WiFi setup I now have the ability to get a lot of telemetry and do stress testing.

Craig suggested a small fan to create circulation within the case. Lots of options...

[Craig Kirkpatrick]

I was conservative with the temperature limits like Chris says.  I had to base it off of the de-rating of the MPD power supply which according to their specs tapers off to nil by 60C ambient.  For folks charging in the Mojave Desert or Las Vegas or parts of California this time of year I think it is pretty easy to bump into the limits, especially charging at 40A like your RAV4 wants.

I did a lot of experiments with adding a tiny 25x25mm fan in the enclosure powered from the +12V and switched on in place of the relays in order to stir the air a tiny bit in the enclosure for the 50A contactor kit.  Simply stirring the air a little bit kept things dramatically cooler. I still use such a fan in my 50A kit that charges my Ford Focus at 6.6kW and it makes a difference.  The problem so far is I'm using a crap fan from China off of eBay and I really want a fan that doesn't make a crazy buzzing noise because it is running out of balance like the POS fan I'm using now.

This motivates me to get samples of the other fan I'm considering.  Summer is upon us and I want folks to find the temperature monitoring a good safety measure and not a nuisance.  Stay in touch with me on this.

-Craig K

[chris1howell .]

In 100F heat 37.7C you are already so close 50C. I plan to try 10C over each setting.

On Jun 8, 2015 1:53 PM, "Craig Kirkpatrick" <crai...@comcast.net> wrote:

I was conservative with the temperature limits like Chris says.  I had to base it off of the de-rating of the MPD power supply which according to their specs tapers off to nil by 60C ambient.  For folks charging in the Mojave Desert or Las Vegas or parts of California this time of year I think it is pretty easy to bump into the limits, especially charging at 40A like your RAV4 wants.

I did a lot of experiments with adding a tiny 25x25mm fan in the enclosure powered from the +12V and switched on in place of the relays in order to stir the air a tiny bit in the enclosure for the 50A contactor kit.  Simply stirring the air a little bit kept things dramatically cooler. I still use such a fan in my 50A kit that charges my Ford Fusion at 6.6kW and it makes a difference.  The problem so far is I'm using a crap fan from China off of eBay and I really want a fan that doesn't make a crazy buzzing noise because it is running out of balance like the POS fan I'm using now.

This motivates me to get samples of the other fan I'm considering.  Summer is upon us and I want folks to find the temperature monitoring a good safety measure and not a nuisance.  Stay in touch with me on this.

-Craig K

On Monday, June 8, 2015 at 1:28:06 PM UTC-7, Chris wrote:

The current temperatures for throttling may be too consecutive we may be able to increase the points by 10c without issue. We have never seen reports of instability due to heat so I think we can raise the limits a bit.

With the WiFi setup I now have the ability to get a lot of telemetry and do stress testing.

Craig suggested a small fan to create circulation within the case. Lots of options...

--
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.

[Craig Kirkpatrick]

Yes, experiments show that it is pretty normal to see a +20C increase in the enclosure without stirring the air a little bit.  The first threshold kicks in at 52C to throttle the pilot amperage in half.  Danny saw also that it restores the original value if the temperatures subside by 3C.  It seems worth trying bumping every threshold up by 5C and if you are brave then 10C.  It is easily edited in the open_evse.h by searching for "TEMPERATURE".  I'd test it myself but we are topping out in the low 80F here in Oregon right now.

I did use the WiFi module to watch the three temperature sensors this morning while fully charging my wife's Fusion again from empty.  In the 50A contactor kit maxing out at what the Fusion wants which is a piddling 15A the temperature maximums were nice and low with outside air temperatures about 78F.  This is dramatically lower than my 30A relay kit with no fan that I showed yesterday reaching 50C on the IR sensor looking at the fuses in that 30A kit.

b

[Andre Eisenbach]

Danny,

Not sure if you can flash on the road, but I just changed the temperature limits on NoSpark.

So next time  you can pull/flash you'll have more conservative limits.

[Craig Kirkpatrick]

I'm tempted to put a water-tight ICSP header on the side of my OpenEVSE enclosures or figure out how I can do it via Bluetooth.  Actually I think I can do it wirelessly with Nick's OpenEVSE II with a slight modification he explained.  I'm weary of opening enclosures to experiment with firmware changes.

[Danny ter Haar]

Of course I took my programmer and my acer chromebook with full debian with me, will pull source and compile later tonight ;-)

[chris1howell .]

Now you just need a WiFi module in your OpenEVSE tethered to your phone. Then we can all watch your temps and charge rate at each stop in real time...

On Jun 8, 2015 3:28 PM, "Danny ter Haar" <from...@gmail.com> wrote:

Of course I took my programmer and my acer chromebook with full debian with me, will pull source and compile later tonight ;-)

[Craig Kirkpatrick]

I think it is time for the OpenEV store to host an Emoncms server      ;-)

[chris1howell .]

Working on it....

[Craig Kirkpatrick]

My enhancement request about giving each ESP8266 module a different Emoncms node is so that OpenEVSE junkies like me with 4  now going on 5 OpenEVSEs can use one firmware for all ESP8266 modules.  Think big.  Heck, maybe the answer is to have the node number be one of the things entered when the WiFi SSID etc is entered.  My idea of basing it off of the mac address is complicated and not always going to work.  

[chris1howell .]

Yep, I will add a node number on the setup page.

[Alan Kirk]

My EVSE went into the temp protection mode on Sunday when it was only 99 in the upper Mojave Desert! 

Paul Gipe had asked to use it as he ventured over to Ridgecrest from Bakersfield and got a partial charge, enough to make it home the next day after a 120v hotel charge. 

Paul will put out a report on his adventure on his web site in the next few days: 

 http://www.wind-works.org/cms/ 

Paul has written about several of his adventures in California with his 2015 LEAF. 

Alan

[Craig Kirkpatrick]

My crazy Peltier cooler idea is starting to not sound so dumb after all.

[chris1howell .]

There is a couple issues with Pelter the first is you have to cut a hole in the enclosure for the hot side. The second more problematic is condensation on the cold side.

If you are going to cut a hole for a simple fan would be enough.

One think to remember is we have never seen a stability issue due to heat. I have been charging 2 EVs in the Mojave Desert since 2011 logging over 100,000 EV miles.

[Craig Kirkpatrick]

Yes.  History has shown that other than some rare isolated thermal runaway events that general warming in the enclosure has not led to a track record of problems or instability.  I'm glad you want to abuse test it and bump thresholds up to prove that raising the thresholds is safe.

[Chris]

Here is my charge session from last night at 32A. It is warm... 100 now in the Mojave Desert.

I was throttled after 1 hour + charging the rise was just about stabilized, another Deg C or 2 and it would have been fine. After the reduction the temp decreased.

[Cory Clemmer]

I wonder about the same type of cooling that most laptop computers use (heat pipes), a heat sink on top of the PSU in the EVSE to absorb case ambient and power supply heat and move it along some copper rods to a heat sink on the outside of the enclosure.

[Craig Kirkpatrick]

Chris, fantastic data.  Try bumping up the thresholds by 5C each.  It will be nifty to see plots in the morning.  

The low default threshold is 52C.  It will be interesting to see if your temperatures plateau lower than 57C.  Or if you download and use the Development branch from Lincomatic you can just comment out #define TEMPERATURE_MONITORING since now the development branch will report temperatures in any state, not just in state C like yesterday.  That way no throttling at all will occur but still you will wake up to a graph of temperature data.

-Craig

[chris1howell .]

What code version has your latest addition for temps in all states? I will bump the limits 10C. and push the current up each night. 

I moved my outside temp probe to the garage so I will graph Garage ambient and EVSE temp. I am stress testing the PS as well with the WiFi module overloading the 5V rail... 

[Craig Kirkpatrick]

Chris, You can download it from Lincomatic's Development branch.  I coded the change about lunch time today so the temperatures are always updated.

The simplest thing for gathering data will be to go into open_evse.h and comment out #define TEMPERATURE_MONITORING.  Then you will get your data no matter how hot it reaches.

https://github.com/lincomatic/open_evse/archive/development.zip

It should be very interesting, downright revealing, in the morning.

-Craig

[chris1howell .]

Ran out of time for tonight... I'll have to try a different test. Reduced charge current by 4A. We will see what the data looks like tomorrow.

[Chris]

Here it the charge at 28A from last night. It throttled after about 2 hours down to 22A. At the rate of 22A temps held steady at 22C over ambient temperature in my garage.

[chris1howell .]

So, here is something interesting...

I just checked back over my energy monitor data the last month since I upgraded to temp monitoring code in May. I have not seen a throttling event at all despite charging at 34A. 

2 things changed in my setup:

-upgraded to from 3.7.4 to 3.7.8

-added Wifi module 

On Tue, Jun 9, 2015 at 7:04 AM, Chris <chris1h...@gmail.com> wrote:

Here it the charge at 28A from last night. It throttled after about 2 hours down to 22A. At the rate of 22A temps held steady at 22C over ambient temperature in my garage.

[Craig Kirkpatrick]

Very interesting.  It would be nifty to see how high the temperature gets if you turn off temperature monitoring.  All of this is making me think we need a button menu to turn it off just like we can turn off other safety checks.  I'll leave it to Lincomatic to do the menu since that is his code. 

Look in open_evse.h for this line and just comment it out.   #define TEMPERATURE_MONITORING  // Temperature monitoring support   

That will turn it off.

Or to bump everything up by 5C you could change this set of lines...

#define TEMPERATURE_AMBIENT_THROTTLE_DOWN 520     // This is the temperature in the enclosure where we tell the car to draw 1/2 amperage.

#define TEMPERATURE_AMBIENT_RESTORE_AMPERAGE 490  // If the OpenEVSE responds nicely to the lower current drawn and temperatures in the enclosure

                                                  // recover to this level we can kick the current back up to the user's original amperage setting.

#define TEMPERATURE_AMBIENT_SHUTDOWN 550          // This is the temperature in the enclosure where we tell the car to draw 1/4 amperage or 6A is minimum.

                                                  

#define TEMPERATURE_AMBIENT_PANIC 580             //  At this temperature gracefully tell the EV to quit drawing any current, and leave the EVSE in 

                                                  //  an over temperature error state.  The EVSE can be restart from the button or unplugged.

                                                  //  If temperatures get to this level it is advised to open the enclosure to look for trouble.

change to this...

#define TEMPERATURE_AMBIENT_THROTTLE_DOWN 570     // This is the temperature in the enclosure where we tell the car to draw 1/2 amperage.

#define TEMPERATURE_AMBIENT_RESTORE_AMPERAGE 540  // If the OpenEVSE responds nicely to the lower current drawn and temperatures in the enclosure

                                                  // recover to this level we can kick the current back up to the user's original amperage setting.

#define TEMPERATURE_AMBIENT_SHUTDOWN 600          // This is the temperature in the enclosure where we tell the car to draw 1/4 amperage or 6A is minimum.

                                                  

#define TEMPERATURE_AMBIENT_PANIC 630             //  At this temperature gracefully tell the EV to quit drawing any current, and leave the EVSE in 

                                                  //  an over temperature error state.  The EVSE can be restart from the button or unplugged.

                                                  //  If temperatures get to this level it is advised to open the enclosure to look for trouble.

[Craig Kirkpatrick]

Yes, I think the other thing that changed is mother nature turning on the heat where you live.  You noticed that temperatures in the enclosure rise about 20C when charging and that jives with what I have observed.  A tiny fan in the enclosure helps when simply making the air something other than standing still around the hotter spots.  I have a few nice fans on order from DigiKey arriving by the end of this week and I'll record some data with and without the fan in my 50A contactor build.

The little wifi module seems to draw 80mA at 5V so that will add a fraction of a Watt of heat.  I think the warmest thing in the contactor kits is the power supply.  Warmest thing while charging in the 30A kits is the pair of fuses.

Yesterday I kept an eye on the TMP007 infrared sensor looking at the J1772 output side of the contactor charging my wife's Fusion at about 14A and it did not get any warmer than the ambient sensors.  I want to repeat that with the Focus charging around 28A to see if that side of the contactor gets warm or not.

[lincomatic]

maybe we should set the temps via rapi as well?

 

From: open...@googlegroups.com [mailto:open...@googlegroups.com] On Behalf Of Craig Kirkpatrick
Sent: Tuesday, June 9, 2015 7:23 AM
To: open...@googlegroups.com
Subject: Re: High temp behavior openevse

 

Very interesting.  It would be nifty to see how high the temperature gets if you turn off temperature monitoring.  All of this is making me think we need a button menu to turn it off just like we can turn off other safety checks.  I'll leave it to Lincomatic to do the menu since that is his code.  I heard from him that he will be away on some travel soon so it may have to wait a few weeks.

 

Look in open_evse.h for this line and just comment it out.   #define TEMPERATURE_MONITORING  // Temperature monitoring support   

That will turn it off.

 

Or to bump everything up by 5C you could change this set of lines...

#define TEMPERATURE_AMBIENT_THROTTLE_DOWN 520     // This is the temperature in the enclosure where we tell the car to draw 1/2 amperage.

#define TEMPERATURE_AMBIENT_RESTORE_AMPERAGE 490  // If the OpenEVSE responds nicely to the lower current drawn and temperatures in the enclosure

                                                  // recover to this level we can kick the current back up to the user's original amperage setting.

#define TEMPERATURE_AMBIENT_SHUTDOWN 550          // This is the temperature in the enclosure where we tell the car to draw 1/4 amperage or 6A is minimum.

                                                  

#define TEMPERATURE_AMBIENT_PANIC 580             //  At this temperature gracefully tell the EV to quit drawing any current, and leave the EVSE in 

                                                  //  an over temperature error state.  The EVSE can be restart from the button or unplugged.

                                                  //  If temperatures get to this level it is advised to open the enclosure to look for trouble.

 

change to this...

 

#define TEMPERATURE_AMBIENT_THROTTLE_DOWN 570     // This is the temperature in the enclosure where we tell the car to draw 1/2 amperage.

#define TEMPERATURE_AMBIENT_RESTORE_AMPERAGE 540  // If the OpenEVSE responds nicely to the lower current drawn and temperatures in the enclosure

                                                  // recover to this level we can kick the current back up to the user's original amperage setting.

#define TEMPERATURE_AMBIENT_SHUTDOWN 600          // This is the temperature in the enclosure where we tell the car to draw 1/4 amperage or 6A is minimum.

                                                  

#define TEMPERATURE_AMBIENT_PANIC 630             //  At this temperature gracefully tell the EV to quit drawing any current, and leave the EVSE in 

                                                  //  an over temperature error state.  The EVSE can be restart from the button or unplugged.

                                                  //  If temperatures get to this level it is advised to open the enclosure to look for trouble.

 

On Tuesday, June 9, 2015 at 7:04:49 AM UTC-7, Chris wrote:

Here it the charge at 28A from last night. It throttled after about 2 hours down to 22A. At the rate of 22A temps held steady at 22C over ambient temperature in my garage.

 

 

--

[chris1howell .]

That would be cool...

Once we get a full blown application or web app, OpenEVSE is going to be a really powerful platform...

All the recent development is really exciting

[Craig Kirkpatrick]

If we set the temps via RAPI then just let the user set the first threshold temp (now is 52.0C ) and then just calculate the other temperatures as 3C increments from that first threshold.  That will make the menu manageable and simple for the user.

[Craig Kirkpatrick]

I added Emoncms dashboards for the temperatures in my two operational OpenEVSEs.  Feel free to take a look and play with the dashboards since you can't hurt them from your end. http://emoncms.org/PlinkerCraig  To support two separate nodes I just changed a line in Chris' ESP8266 code to hard-code the second node number like this:

// We now create a URL for OpenEVSE RAPI data upload request

  String url = "/input/post.json?node=2&json={";

[Craig Kirkpatrick]

I need to take my Ford Focus for a long drive to run the battery down so I can charge for >3 hours.  The plot below was less than 30 minutes at 28Amps on my 50A contactor kit with the little eBay fan inside the enclosure.  The TMP007 infrared sensor is watching the J1772 output lugs of the 50A contactor.

[Chris]

 Here is my charge from last night. The ambient temp was a couple degrees cooler, just enough that throttling would not have happened with the old threshold as the max reached 51.7.

[Craig Kirkpatrick]

I like seeing how the temperature reached a nice plateau rather than continuing slowly upward.  I gathered some data this morning also.  This was my Ford Focus charging at about 28A for over an hour.  The temperatures are slightly affected by morning sunshine on my EVSE and J1772 cable.  Again the Infrared reading is a TMP007 sensor in my enclosure attached to the lid and staring at the J1772 side of the contactor.  I was also happy to see my temperatures reach a nice plateau like yours did.

.

[chris1howell .]

Yes, the plateau is very even over 3 hours on my charge. Looks great I will up the amperage and see what the change is.

I expect a slight increase in the delta between Ambient outside temp and enclosure.

We have a few heat sources that will be the same no matter the charge current.

AC_Test 1-2 watts of heat

Board, PS, load resistance

Relay coil

Really the only variable is increased heat due to current. 

I think the key is to find a reasonable value slightly above the plateau (within the limits of the hardware) and at a decent ambient Outside air temperature... That way we can determine normal and not and limit temperature below the point of instability.

[Craig Kirkpatrick]

Chris, I like your plan.  I took it upon myself to add the menu item to turn on/off temperature monitoring.  I'm in the middle of testing my code right now.  I'm beginning to see how much flash that menus eat up, so I might hold off on adding a menu to adjust the temperature threshold and instead we just determine a more reasonable threshold like you are planning to do.

[Craig Kirkpatrick]

Chris, Alan, Danny, and others bumping into the temperature thresholds, I'd like you to bravely test this development branch I labeled D3.9.3.1

https://github.com/craigk46/open_evse/archive/development.zip

This increment adds a menu item in the setup menu to turn off Temperature Checking.  I also went ahead and raised the existing thresholds by 3C so that instead of the first threshold kicking in at 52C it will kick in at 55C.

I tested the menu item actually turns on and off the temperature monitoring by employing the #define TESTING_TEMPERATURE_OPERATION in open_evse.h which makes it simple on the bench to run into testing thresholds.   I trust it works but otherwise I hope it is hard for you to reach the 55C threshold.

Let me know if you have any feedback.

-Craig K.

On Wednesday, June 10, 2015 at 10:53:11 AM UTC-7, Craig Kirkpatrick wrote:

Chris, I like your plan.  With Lincomatic away on some travel I took it upon myself to add the menu item to turn on/off temperature monitoring.  I'm in the middle of testing my code right now.  I'm beginning to see how much flash that menus eat up, so I might hold off on adding a menu to adjust the temperature threshold and instead we just determine a more reasonable threshold like you are planning to do.

[Chris]

I got some good data last night.

It was hot.

I needed a long charge.

I upped the current by 6A. I will go to a full 40A next.

The 55C limit would have been fine for this test (mine are at 62 right now for testing).

[Craig Kirkpatrick]

That is really nice data.  Again I'm very happy to see the temperature reaching a plateau.

After watching ambient and infrared temperatures in my two OpenEVSEs that are reporting data to Emoncms I'm developing theories of how the OpenEVSE gets warm and how and why it doesn't just get warmer and warmer during a long charge.  My theory is that the best heat sink into and out of the enclosure is the AC mains wires and the J1772 wires themselves.  I base this a bit off of watching the temperatures of the J1772 wires with the infrared sensor in my 50A contactor build.

I got three nice 25mm fans that will run from 12V DC.  I'll use one to do what I planned which is to stir the air in an enclosure to see how much it aids in overall temperature performance.  I'm tempted to take an old aluminum enclosure and use the other pair of fans as forced air in and forced air out of the enclosure just as an experiment to see what it does for temperatures.  I'm not advocating ventilating the enclosure since I think it is very important to keep dust and bugs and water out of the enclosure.  A fully enclosed EVSE is best in my opinion.

[Chris]

Last Night it was really hot in the garage.  I went up to 40A and was throttled... Not by OpenEVSE but by the Model S. Unfortunately, little is know about how Tesla's reduction works other than it cuts current by 25%.

Here are the graphs.

[Craig Kirkpatrick]

Interesting that temperatures reached about 58C and the OpenEVSE stayed happy. This is what you were saying about the specifications for the MPD supply probably being way too conservative. 

--
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+unsubscribe@googlegroups.com.

[chris1howell .]

The weather here is super hot so I will keep getting data. I saw 105 on the thermometer at lunch time.

[Craig Kirkpatrick]

When I lived in Lancaster, CA in the late 70s I remember 114F being about the most extreme high temperature and it never seemed to exceed 114F. Somehow next to that humongous dry lake where you live it seems possible to poke even higher than 114F.

[Chris]

Here is last night... It was very hot yesterday (105F) and garage did not cool too much.

Charged at 40A and reached 60C. 

The current temp in my idle charging station is 41C or about 105F. Looks like a pretty very consistent 20C rise.

[Craig Kirkpatrick]

Yup. 20C rise is what I was expecting. Wonderful to see OpenEVSE tested at 60C sustained and no problems. I'll repeat what I said last night that MPD was conservative in their specifications of their power supply.

[chris1howell .]

Yes, and I am abusing mine. I need to remove the 30ma load resistor. I am a bit over the Max on the 5v rail.

Now I just need to charge mid day in the 100+ heat and see if I can reach a point of instability (or Shutdown).

[Craig Kirkpatrick]

With the ESP8266 module reporting temperatures you'll be able to see any problem I think if it occurs.  Adding a data feed of EVSE state A/B/C could be revealing also.  With the old ESP8266 firmware that you created I added a feed for fault counters, and it is easier now to add feeds with Lincomatic's enhancements to your code I think. My fault counter stuff is obsolete code now.

[chris1howell .]

I plan to add just about everything to feeds. My vision is apps can hit the data.openevse.com server and send info up and down...I still need to look at lincomatic code.

Found an interesting piece of info...

From the MinMax data sheet:

Over temp protection at 90℃ (automatic recovery at 67℃)...

--

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.

[Craig Kirkpatrick]

That is a lot of margin from MinMax!  

Now its time to consider if another component beyond the power supply is the temperature limiting item in OpenEVSE.  It seems to me that anything north of 70C is asking for trouble, for problems with MTBF.

Tomorrow I have my Peltier module arriving.  Probably it will take me until July 4th weekend to properly integrate it.  I've been experimenting with a tiny module just for fun and to learn the principles of the heat pump.  It is extremely interesting to me.

The last bit of ESP8266 code from Lincomatic made it easier for you or me to add any feed via RAPI from the ESP8266 module, look in your emails from him from about Friday.

[vegasbrad]

I'm willing to add some hardware and participate in the temp data gathering. It very hot here in Vegas now, were seeing 100 plus degree days on a regular basis. I'm have been running a portable openevse built successfully on FW v3.3.4  On two builds running v3.7.8 I ran into trouble, One at my house during testing and one at another EV drivers house, I had to downgrade both to 3.3.4 and haven't had a problem since. At that point i have had to stay with 3.3.4 for the rest.

I like the safety of the temp control, just need it to be happy running in the dessert :)

On Tuesday, June 16, 2015 at 7:33:26 PM UTC-7, Chris wrote:

[Craig Kirkpatrick]

Brad,

The development branch of the code has a button-menu choice to disable the temperature monitoring.  

https://github.com/lincomatic/open_evse/archive/development.zip

Alternatively you can edit the code since I know you are familiar with that.  Look in open_evse.h to find the temperature thresholds and you could just add 5C to the existing thresholds.  I'm pretty satisfied from Chris' observations of high temperature reliability to bump the base threshold of 52C and the other thresholds up by 5C.

I want to make it so that you can bump the threshold up or down via the button menus.  I just haven't done that coding yet, distracted by other things lately.

I added the ESP8266 Huzzah WiFi module to both of my operational OpenEVSE kits and I've been recording temperature and kWh for about a month now to the cloud.  You can poke around at my dashboards and see what temperatures I've been seeing correlated to the Wattage.  You can zoom a graph by Day or Week etc. It is pretty interesting and I reached 50C as a high a few days ago, today is a cool day in Oregon.  Click the various dashboards near the top.  http://emoncms.org/PlinkerCraig   Chris now has a server running to host the cloud data and I should switch over to his server.  I just want to figure out how to do that without losing a month of useful data.

On Monday, July 6, 2015 at 10:29:33 PM UTC-7, vegasbrad wrote:

I'm willing to add some hardware and participate in the temp data gathering. It very hot here in Vegas now, were seeing 100 plus degree days on a regular basis. I'm have been running a portable openevse built succesfully on FW v3.3.4 On two builds I ran into trouble, One at my house during testing ans one at a guys house, I had to downgrade both to 3.3.4 and haven't had a problem since. At that point i have had to stay with 3.3.4 for the rest.

[Matthew Blackler]

Seeing very interesting behaviour with sub-zero (F) temperatures in MN.  Basically, the unit "sees" 255.5 F once it gets down really low and therefore cuts in the thermal protection.  Looks like there might be some bit shifting going on, that doesn't take into account negative numbers or the like.  This is happening on 3.9.0.

[chris1howell .]

Yeah, it looks like quite a few folks are experiencing the sub zero "overheat". Temperature monitoring can be disabled in the menu while it is so cold as a workaround until a firmware fix can be released.

[ssst...@gmail.com]

Chris, I have 3.7.8 and am experiencing this problem every day it gets below 0. There doesn't appear to be any option to disable the temp controls on my OpenEVSE. Please advise.

[ssst...@gmail.com]

OK, I see above there is an option only in a development branch. This makes me rather nervous as 1) it's a development branch 2) I've never applied an update to my OpenEVSE so I'm not familar with the process and 3) I am afraid I could end up with a non-working L2 EVSE. In the below-zero temps it's kind of critical for my daily driver. 

Thanks.

[Danny ter Haar]

On Monday, January 18, 2016 at 9:00:07 AM UTC-8, ssst...@gmail.com wrote:

OK, I see above there is an option only in a development branch. This makes me rather nervous as 1) it's a development branch

I have been using development branches most of the time I own openevse (about 5 years now) and I have not yet once had that my car didn't have a charge when I needed it.

 

2) I've never applied an update to my OpenEVSE so I'm not familar with the process and

Only one way to get familiar: just do it (tm) ;-)

It either works or it doesn't unless power drops while programming, but I think it will even survive that.

 

3) I am afraid I could end up with a non-working L2 EVSE. In the below-zero temps it's kind of critical for my daily driver. 

Worst case, you can use the trickle charger that came with the EV while you wait for a solution ?

[chris1howell .]

3.10.3 in the stable branch has the option to disable temperature monitoring. There is a pre-compiled HEX file at https://github.com/OpenEVSE/

[ssst...@gmail.com]

Danny, thanks for the confidence builders. ;)

Chris, thanks--I will see if I can borrow a friends' programmer and go with the 3.10.3.

[vegasbrad]

This is a very nice guide Chris provided: http://openevse.dozuki.com/Guide/How+to+Load+OpenEVSE+Firmware+%28WinAVR%29/7?lang=en
Just follow that and you can program a HEX file fairly easily.