False Positives With FW 5.8.4 Charge Current Reduction?

[pbrulott]

I posted this in the Firmware 5.8.4 thread before knowing this one existed:

Reduced current and Ring of death struck me for the first time; really unpleasant experience. Some of it is related to 5.8.4 for sure. There seems to be 2 or more things at play here, it's not only the -25% current draw but also, in my case, charging stopped every now an then with the red charging port ring every time.

Story:

Left my place with 291km of range, arrived at destination - Bromont - with 123km (a 102km ride took 168km of range)
Stayed there 2 days, needed 175km of range to be comfortable to come back
I plugged at 10am the first day on a 110V outside outlet with my usual extension cord (gage 14, 50 feet) - I plugged many times with it without issue
After the ski, at around 5pm, I noticed that the car is not charging anymore and has 108km of range ... => red ring
I pulled the UMC and plugged back a couple of times, I reset the screen before it works again, checked all the connections and all looked good (I was getting messages like "bad wiring or extension cord, reduced current and then unable to charge, contact Tesla service" alternatively depending if it was charging or if charging was stopped; it finally charged from 6pm to noon the next day at 4km/h which was ok for what I needed (the last time I checked before going to ski I was at 164km of range)
Called Tesla hotline and they were not helpful at all; they said the Service Center would call me the next day which they didn't
The next day, I went to ski and checked the app while skiing => charging had stopped
Came back to the cottage and red ring again with 160km of range
Needed to leave so I plugged it back for maybe 15 min and when it started to work (after a couple of unplug/plug) I got 0km/h of charge but range started to increase (it was -17C/0F outside) so battery must have been heating before charging.
Left with 164km of range, put the heating on at minimum (64F) and drove at 55 mph
I finally reached destination with 10 miles remaining but quite a bad experience of charging in a remote location
Not something my dad or my wife would have been able to deal with

Anybody knows what could have happened. I sent the same message to the Service Center. I don't like this latest update :-(

_____

forgot to mention that I charged using the same extension cord and UMC when I arrived home and was getting 6km/h no messages...

PB

 

[Electric Machete]

After the most recent software update, my mobile connector stopped working. I took it to Tesla in Dania Beach, FL and they said the connector was fine and they have had issues with the UMC since the update. I returned home and it was working again, but I am only charging at 30amps when I had been at 40amps for the last year prior to upgrade. I assume they will sort this out with the next update.

 

[InTheShadows]

I am currently charging at 30 amps. 238v, first time I have noticed the reduced charging rate. The battery is warm too so it is not reducing it for that reason. I'll check the voltage at the breaker when I head out again.

 

[kota23]

From an electrical standpoint the reduction in charging amperage makes a lot of sense. Voltage can vary widely depending on time of day and what else is loading the system. Until electrical code catches up and creates better standards for EV charging then we are faced with what we have now, Tesla reacting as best they can. The fact that Tesla is reacting when problems arise and not just standing idly by is a positive. I'm sure this will not be the last firmware update to deal with charging. Recently the National Electric Code added arc-fault breaker protection for all new home branch circuits and any remodel work to electrical home circuits now requires installation of arc-fault breakers. Garages are still excluded, but I would say requiring an arc-fault breaker for EV charging circuits is not far off. Will be big money for GE, Square-D, Eaton, etc.
http://www.geindustrial.com/cwc/Dispatcher?REQUEST=PRODUCTS&id=cb-qafci

 

[mep]

It would be great if you reported the Voltage before the current ramped up, and then the Voltage and current after the current has ramped up. Maybe a Voltage at 40 Amps and then after the fold back to 30 Amps. From the Voltage drop at load currents it is possible to get a rough estimate of the resistance in the installation and see if that is a threshold for the charge reduction.

It took me a few days of testing to catch the car in the act, so to speak. I am charging at a NEMA 14-50 at 40A. Industrial power provided by ComEd in the Chicago area is 208V.
Without load it shows 214V. When charging starts I read between 198V and 201V, it fluctuates a bit. It always takes some time (between 10min and 1 hour) before I get that error message about faulty wiring or use of extension cord and the car reduces charging to 30A. Once it charges at 30A I cannot manually increase to 40A. I have to unplug and reinsert the charging cable. Then I still need to manually increase charging back to 40A until the next "event". Today I was sitting in the car for a full hour waiting for the moment. What happened is that the voltage for a very short moment fluctuated down to 196V. I was watching the screen at that moment and in that split second I got the error messages and the car went down to 30A. So in case of a 208V line and 40A charging it appears that the low limit is set at 196V. Not sure whether it is a faulty line. In fact power provided by ComEd might fluctuate. However, from reading this thread it could also be the charger in the car? What is the email address that people use to contact Tesla to look at the logs? I would like to get in touch with them.

 

[dirkhh]

It took me a few days of testing to catch the car in the act, so to speak. I am charging at a NEMA 14-50 at 40A. Industrial power provided by ComEd in the Chicago area is 208V.
Without load it shows 214V. When charging starts I read between 198V and 201V, it fluctuates a bit. It always takes some time (between 10min and 1 hour) before I get that error message about faulty wiring or use of extension cord and the car reduces charging to 30A. Once it charges at 30A I cannot manually increase to 40A. I have to unplug and reinsert the charging cable. Then I still need to manually increase charging back to 40A until the next "event". Today I was sitting in the car for a full hour waiting for the moment. What happened is that the voltage for a very short moment fluctuated down to 196V. I was watching the screen at that moment and in that split second I got the error messages and the car went down to 30A. So in case of a 208V line and 40A charging it appears that the low limit is set at 196V. Not sure whether it is a faulty line. In fact power provided by ComEd might fluctuate. However, from reading this thread it could also be the charger in the car? What is the email address that people use to contact Tesla to look at the logs? I would like to get in touch with them.

So NEC (national electric code) suggests a maximum drop of 5%. As it happens 197V still rounds to 5% drop, but 196V rounds to a 6% drop. Could be coincidence.
If that were the threshold, I'd assume that on a 240V circuit the cutoff would be at 226V - I have never seen anything nearly as low in my data... but of course it could be very brief fluctuations that trigger this that my logging might have missed (as that samples only once a minute).

 

[InTheShadows]

So NEC (national electric code) suggests a maximum drop of 5%. As it happens 197V still rounds to 5% drop, but 196V rounds to a 6% drop. Could be coincidence.
If that were the threshold, I'd assume that on a 240V circuit the cutoff would be at 226V - I have never seen anything nearly as low in my data... but of course it could be very brief fluctuations that trigger this that my logging might have missed (as that samples only once a minute).

This is good to know. This is expected behavior then. All it takes is a huge load in the building or in another building to draw a startup load and dip the voltage for a second to trip it out.

In my case today I could manually raise the voltage back up. I just started using visible tesla two days ago to try and track when it dips. I looked at the log and my car had been charging at 30 amps both days. It stayed at 40 once I dialed it back up again.

 

[Larry Chanin]

When I first charged after the update my car charged as it always has at 40 amp.

I just checked my charging and it had dialed itself down to 30 amps without providing any warning message.

I proceeded to manual dial it back up to 40 amps and after watching for about 10 minutes it continued to charge at 40 amps. The no-load voltage was 246. The full 40 amp voltage was 243.

I'm inclined to consider this a problem especially since the car didn't provide an error message after dialing back the current.

Larry

 

[Forty Creek]

My HPWC has been consistently dialed back by 25% over the past week and displaying the 'extension cord in use or bad wiring' message. I manually lowered the amps a couple of ticks (80 to 77) to see if that would make any difference. It didn't. So next step will be to call Ontario Hydro to have the line checked at the street. We live in a somewhat rural setting and our neighbors had a problem with a load unbalance in the line a few months back. Perhaps this is our issue as well.

 

[Rainbow]

I think it is wrong for people to assume they are getting false positives. Have you monitored the voltage and amperage? Have you monitored it properly? Do you know what to look for? Do you have the right monitoring equipment?

While there should be limits to the input voltage, the most common, signature of a connection failure is not typically slow or slight voltage/amperage changes. I don't know what type of detection Tesla uses, but the I think charging detection should look for the signature of arcing. Very fast and significant voltage changes.

If you are getting fault messages or if your charger is charging at a reduced rate; I would strongly suggest inspecting your cords and connectors for signs of overheating and arcing. If you are getting fault messages or if your charger is charging at a reduced rate you should contact Tesla.

If Tesla is has advanced detection: It's possible to get false positives from other devices on your service or possibly on neighboring services. High amperage devices especially with coils, could give false positives, especially if those devices have faults. Air-conditioners, heat pumps, arc welders, refrigerators, often emit an arc signature. Especially if they have a have a bad capacitor, failing coils, bad or dirty contacts.

You also might get more false positives in rural setting, using extension cords, long runs, and or if your service is marginal capacity.

By assuming false positives, you might be over looking potential danger.

BTW "you" in this context, is a figure of speech that can mean anyone.

 

[wcalvin]

This is a typical charging scenario measured when charging via my J1772 the day after Christmas. I'm measuring the current and voltage at the wall. See also http://www.teslamotorsclub.com/show...uring-the-Total-Energy-Consumption-of-Your-EV

(Axes are scaled so that the swings are more visible.) Total voltage swing during the charge is about 0.7% of the nominal voltage and current swing is about 0.3% of my nominal 30A charging current. The voltage drop during the charge is about 1.5%. I haven't had any current reductions with this level of "noise" in the charging supply.

View attachment 39731

That second voltage drop at 01:15PM shows no change in current, nicely illustrating a sudden source resistance increase, exactly what I was mentioning earlier at #24.

That resistance increase is what they need to detect, not voltage drop per se. Usual reason is likely a decrease in plug-receptacle surface contact area, as plug resettles into a new position slightly out from earlier. Usual setup for arcing.

 

[InTheShadows]

I'm running visible tesla all the time now so I can see what is going on.

I noticed that once it gets reduced it stays reduced and you need to manually turn it back up. A better behavior might be to automatically dial back up 1 amp at a time every 10 minutes. Then reducing again if it runs into problems on the current charge cycle and sending a notification to the owner.

 

[Cottonwood]

That second voltage drop at 01:15PM shows no change in current, nicely illustrating a sudden source resistance increase, exactly what I was mentioning earlier at #24. That resistance increase is what they need to detect, not voltage drop per se.

Or some other load coming on. I have an ETS (Electro Thermal Storage) heater in my house that charges when the rates are low. When it turns on all of its elements to heat the ceramic bricks, it can pull almost 50 kW. The off-peak rate starts at 11pm. Look at the graph below. The ETS coming on with a 50 kW load at 11 pm caused a drop of about 6 Volts out of 240 Volts or a 2.5% Voltage drop that had nothing to do with a resistance change; it was just increased current in a common resistor (HV lines, transformer, and feeds to my house).

The power is only that power going to my MS, but the Voltage is measured in the breaker panel feeding the HPWC. See Charging while preheating cabin w/ HVAC for more detail on the MS power draw.

Update:

I went through the paths for the power into my house. Even though the load (its actually 46kW) is pretty large, my electrical connections are very stiff.

My house is the only user on a 100 kVA transformer fed by a 7.2 kV line. There is less than 20 feet of cable from the transformer to my main distribution panel. I am not sure what the wire size is, but I bet it is probably something on the order of 1000 MCM Al wire (1" diameter Al!). These wires go through current transformers that feed the meter, so there is no loss at the meter. Those big ass wires feed a big ass 600 Amp breaker. The breaker is especially big because it has to handle the short circuit current of that 100 kVA transformer and not weld itself in a connected state. This distribution panel has 2, 200-Amp circuit breakers that each feed a 200 Amp sub panel near the ETS heater. The ETS heater is fed by 4, 60 Amp breakers, two in each of the 200 Amp panels. Also back in the first 600 Amp distribution panel is a 400 Amp breaker that feeds my garage and other parts of the house above the garage. The HPWC is fed by a 100 Amp breaker in the 400 Amp garage panel.

All of this is to say that the only common path is the HV line, the transformer, the 20 feet of giant wire to the 600 Amp distribution panel, the 600 Amp breaker and the big bus on the distribution panel for the ETS heater to cause voltage swing to my HPWC. That is a lot of heavy duty wire, etc, and the 46 kW load still gives a 6V, 2.5% drop.

Whatever Tesla is using as a detection algorithm, it must take into account the fact that other loads/users of the grid will cause Voltage swings of several percent.

 

[tezco]

That second voltage drop at 01:15PM shows no change in current, nicely illustrating a sudden source resistance increase, exactly what I was mentioning earlier at #24.

That resistance increase is what they need to detect, not voltage drop per se. Usual reason is likely a decrease in plug-receptacle surface contact area, as plug resettles into a new position slightly out from earlier. Usual setup for arcing.

Here's a week's worth of voltage measurements. (Voltage axis not nearly as magnified as in the previous post.) Fairly active swings out here on the farm. I don't think I have a load at the house that would explain the voltage drop over such a long interval, and I have 400A service, so it's probably happening outside of my neighborhood. I do have solar, but insolation values for that time period were also steady.

I suspect that the reason that there is no change in current despite a drop in voltage is that the MS is regulating the current input based on my setting of 30A max.

 

[cinergi]

I just got nailed by this twice tonight on my newly-installed HPWC and 200 amp service. Power in my area sucks -- I usually start out around 230V and then under load I see 220-218 ... the neighborhood has a fair amount of fluctuation so this evening I observed it fluctuate from 218 to 213. And at some point it decided to reduce the current. I've got a message into Tesla about it.

 

[Andrew]

I'm encountering the problem on my home NEMA 14-50, and am in communication with Tesla (waiting to hear back from them after they review my logs).

I haven't voted in the poll yet, because I'm not sure if it's a "false positive," or just a "positive." Though I expect our wiring should be just fine -- the entire circuit is brand new, with a new panel/meter/weatherhead run just for the car. Additionally, the drop from the utility pole was replaced about two years ago. It's about an 80' run from the breaker to the garage outlet, and our electrician used 6 AWG wire.

Will report back once I learn more...

 

[Lloyd]

I had 198V at a chargepoint station yesterday with full amperage (30A) yesterday with no degradation from the car for three hours in San Jose.

 

[laalan]

Home 14-50 lowers to 30a every few days. No rhyme or reason.

 

[notailpipe2112]

I saw this happen on a Sun Country CS90 charger at Hotel Mortagne in Boucherville, QC in the past two weeks. The scenario for me was that I was driving out east over the holidays and I stopped to use this charger before Christmas, and before the software update. I believe I was getting 69 amps with about a 200-ish volt situation. Power was consistent when I first used it and I think I was getting 60 km+ charge rate per hour. At the time, I never saw a warning on this charger that it needed to be limited by 25%, or anything like that, and my charge rate was consistent the whole time.

On the way back I stopped at the same charger, and after the software update, and around the same time (dinner hour) looking for the same amount of charge (200 km worth). This time I plugged in and it was fine for the first 5 minutes and then it suddenly gave the limiting warning that is new to the new software. The net result was that I got limited to only 50 km charge rate per hour, and it meant I was waiting much longer (with my whole family and dog) in Boucherville. Soon as I saw the message, I knew it had to be related to the software update. I was cursing Tesla, I have to admit, because I knew this wasn't an issue with this charger just 5 days prior but now the car/software suddenly deemed this charger unfit. Bugger! It was my fault for taking a software update in the middle of my trip.

On the plus side, I have note seen this issue at my home or cottage, which both have a clean 40 amp/240 volt set-up consistently providing a charge rate of 45 km per hour for my car.

 

[neroden]

I was cursing Tesla, I have to admit, because I knew this wasn't an issue with this charger just 5 days prior but now the car/software suddenly deemed this charger unfit. Bugger! It was my fault for taking a software update in the middle of my trip.

Well, yes, and I was smart enough to NOT get the software update in the middle of my trip. But the underlying problem is entirely Tesla's fault. Tesla's new software algorithm for preventing charging is crap. The Sun Country CS90 is obviously just fine -- it just runs on a lowish voltage. I plugged in at four Sun Country CS90 locations with voltage patterns like this during my road trip. I can't imagine how ruined my trip would have been with the 5.8.4 "upgrade" to the firmware.

If Tesla has any sense, Tesla will simply revert this firmware degradation ASAP. Time enough to figure out a better algorithm without breaking everyone's road trips.