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

PowerWall installation mess: Need advice on how to get it resolved

valleydude

Member
Dec 21, 2019
24
3
California
Thanks for the pictures. Per the markings on the dead front, the only "non-spare" breakers are a 90A for the A/C, a 90A for the GW, and a 40A for the EVSE. So two conclusions:

(1) Per the labeling, everything in the outdoor panel that you wanted backed up has been moved out to the newly installed Backup Panel. So if you believe otherwise, either (a) Tesla has mislabeled the dead front, and one of those breakers marked spare is still feeding a circuit or (b) the circuit that you are expecting to be currently backed up (because it had been in the main panel), but you find it isn't, you are mistaken and it actually originates in the garage panel. You can easily distinguish between the two possibilities by turning off all the "spare" breakers (right now the group at the upper left is off, but the two spares on the upper right are on, as are the spares at the bottom). If some circuit gets shut off, you know the issue is (a); but if nothing changes, then the issue is (b).
That is correct. All the loads on the main panel outside the house that I was expecting to be backed up having been relocated to a new panel they installed inside the garage, except for the EVSE.
(2) The above breakers total 220A, so Tesla has left you with a technical code violation. It's not actually a safety rule, as the NEC "sum of all breakers rule" is a bit stricter than physics requires. But it may be useful ammunition for negotiating with Tesla to get someone back to fix the mess they left.
This is good know.
Oh, and those stickers at the bottom that say "Solar Breaker Do Not Relocate," but the breakers also say "Spare" on the Dead Front, those stickers are no longer accurate (assuming Tesla moved the solar to the internal panelboard in the Gateway as it was supposed to). So you can remove those or cross them out.
Yes. They should have removed the stickers. I will do it.
Cheers, Wayne
Now the real issue is, I am not able to get past the Project Advisor and talk to a design engineer to explain the situation. Whatever I tell the PA either verbally or communicate over email, seems to be getting lost in translation.
All I heard from the PA is that their design engineer does not want to make any changes and it can only be solved by adding an additional PowerWall, which even if I wanted, they will not sell it to me :(

cheers and thanks for your help
 

wwhitney

Member
Nov 2, 2017
968
1,317
Berkeley, CA
On the EVSE, I recommend against backing it up if your EV is not a Tesla; otherwise, if you have a power failure at night while your EV is charging, it will merrily go on charging and deplete your Powerwalls.

As to the PA, I have no idea what the best approach to take with Tesla would be. My instinct would be to try something along the lines of "per the contract, all loads except the AC unit were to be backed up. Per the equipment installed, only half of the non-AC loads are backed up; half are not. Please reconfigure and reinstall as necessary to back up all non-AC loads." Of course, if the part about backing up everything except the AC is strictly an oral agreement, it will be harder to enforce.

And rereading the PA's response in the OP, they are just showing a lack of imagination in how to install the system.. If you had over 200A of loads you wanted backed up, or if the (2) panels were far apart and each contained both loads to be backed up and loads not to be backed up, then it would be impossible or challenging to back everything up without using 2 Gateways. But since the only non back up load (or two) is right there at the main panel, it's not hard to configure the Gateway and feeders to back everything else up; my proposal is just one option.

[Here's another option that Tesla might like better, as it doesn't involve messing with the 100A and 200A main breakers in your "split bus" panel. Use the 100A main breaker to supply your AC (might have to set a disconnect right next to the main panel with a 90A breaker.). Move the EVSE over to your new back up panel. Put a 4 position 200A branch breaker into the 200A distribution section of your main panel (that breaker is listed on the main panel label as allowed). The 200A branch breaker supplies a 200A feeder to the GW, and from the GW to the new back up panel. Put a 100A breaker in the new back up panel to refeed the garage.]

Cheers, Wayne
 
  • Like
Reactions: valleydude

valleydude

Member
Dec 21, 2019
24
3
California
On the EVSE, I recommend against backing it up if your EV is not a Tesla; otherwise, if you have a power failure at night while your EV is charging, it will merrily go on charging and deplete your Powerwalls.
I agree. My EV is a Tesla M3, which I believe will communicate with PowerWall and will not drain it completely during a power outage.
As to the PA, I have no idea what the best approach to take with Tesla would be. My instinct would be to try something along the lines of "per the contract, all loads except the AC unit were to be backed up. Per the equipment installed, only half of the non-AC loads are backed up; half are not. Please reconfigure and reinstall as necessary to back up all non-AC loads." Of course, if the part about backing up everything except the AC is strictly an oral agreement, it will be harder to enforce.
I don't see it mentioned in the contract in writing that all loads other than the ACs should be backed up. It is a verbal agreement. So, I'm out of luck on that front.
And rereading the PA's response in the OP, they are just showing a lack of imagination in how to install the system.. If you had over 200A of loads you wanted backed up, or if the (2) panels were far apart and each contained both loads to be backed up and loads not to be backed up, then it would be impossible or challenging to back everything up without using 2 Gateways. But since the only non back up load (or two) is right there at the main panel, it's not hard to configure the Gateway and feeders to back everything else up; my proposal is just one option.
I'm not trying to exclude any loads from the sub-panel in the garage and they have already excluded the heavy loads from the main panel. So, as you say, they should be able to backup everything except the heavy loads.

The sub-panel in the garage would be 15 - 20 away from the panel. Would that be a problem?

[Here's another option that Tesla might like better, as it doesn't involve messing with the 100A and 200A main breakers in your "split bus" panel. Use the 100A main breaker to supply your AC (might have to set a disconnect right next to the main panel with a 90A breaker.). Move the EVSE over to your new back up panel. Put a 4 position 200A branch breaker into the 200A distribution section of your main panel (that breaker is listed on the main panel label as allowed). The 200A branch breaker supplies a 200A feeder to the GW, and from the GW to the new back up panel. Put a 100A breaker in the new back up panel to refeed the garage.]
I like this option. I will try to reason with Tesla PA and see how far it will get me.
Cheers, Wayne

Thanks once again for your insightful responses, Wayne. Much appreciated.
 

wwhitney

Member
Nov 2, 2017
968
1,317
Berkeley, CA
The sub-panel in the garage would be 15 - 20 away from the panel. Would that be a problem?
No.

Also, if Tesla won't give you satisfaction, any knowledgeable electrician could make the changes I proposed, it would just cost you some number of kilobucks. And finding a knowledgeable electrician could be hard, a significant number might be unwilling to modify what Tesla has done.

Cheers, Wayne
 
  • Like
Reactions: valleydude

valleydude

Member
Dec 21, 2019
24
3
California
No.

Also, if Tesla won't give you satisfaction, any knowledgeable electrician could make the changes I proposed, it would just cost you some number of kilobucks. And finding a knowledgeable electrician could be hard, a significant number might be unwilling to modify what Tesla has done.

Cheers, Wayne
@wwhitney Wayne, it took me a while to fully understand the last solution that you proposed. I tried translating it into a block diagram. According to my interpretation of your proposal, the orange colored blocks are what need to be installed and the corresponding electrical connections need to be made. Please ignore my lack of familiarity with electrical wiring conventions and let me know if my block diagram captures accurately what you suggested.

thanks.
 

Attachments

  • Screen Shot 2021-07-08 at 9.29.15 PM.png
    Screen Shot 2021-07-08 at 9.29.15 PM.png
    147.8 KB · Views: 26

wwhitney

Member
Nov 2, 2017
968
1,317
Berkeley, CA
For the most part, yes. BTW, that type of block diagram is called a one line diagram and you've executed the format well. Some minor terminological notes: in the main panel, call the lowest block "distribution section" rather than "distribution panel"; also the meter is in the main panel enclosure, so it would typically be a small box in the upper left of the big box. Two substantive comments:

1) The need for the new 90A disconnect depends on the details of what's downstream of the current 90A breaker in the main breaker panel. There's a small chance that you have a single AC unit, which is marked with, say, MCA 80A (minimum circuit ampacity), MOCP (maximum overcurrent protection) 100A, and the circuit conductors have an ampacity of 100A already (say #3 Cu), and you could eliminate that 90A disconnect. Not very likely, as the current breaker is 90A, but a detail to be checked.

2) The rendering of the panel downstream of the Powerwall isn't accurate. It wouldn't have two 100A breakers in parallel as your diagram currently suggests. (The way your main panel does that is not common.) It would have a new 200A main breaker (in the main breaker position, rather than the backfed 100A branch breaker you currently have) along with the upsized feeder from the main panel to the Gateway to the new 200A main breaker. Which should work as that panel is supposed to have a 200A main bus already. All the existing relocated breakers on its distribution bus would remain unchanged. And a 100A breaker would be used on the distribution bus to supply the panel in the garage. Which, since the existing backfed 100A branch breaker won't be used anymore, could be the exact same breaker in the exact same location, just changing what wires go it.

Cheers, Wayne
 
Last edited:
  • Helpful
Reactions: jjrandorin

valleydude

Member
Dec 21, 2019
24
3
California
For the most part, yes. BTW, that type of block diagram is called a one line diagram and you've executed the format well. Some minor terminological notes: in the main panel, call the lowest block "distribution section" rather than "distribution panel"; also the meter is in the main panel enclosure, so it would typically be a small box in the upper left of the big box. Two substantive comments:
Updated the one line diagram to reflect your comments. Please take a look at the attached picture.
BTW, in my case, the meter and the main panel are adjacent to each other and not on the same panel.
I've attached the pictures for my installation.
1) The need for the new 90A disconnect depends on the details of what's downstream of the current 90A breaker in the main breaker panel. There's a small chance that you have a single AC unit, which is marked with, say, MCA 80A (minimum circuit ampacity), MOCP (maximum overcurrent protection) 100A, and the circuit conductors have an ampacity of 100A already (say #3 Cu), and you could eliminate that 90A disconnect. Not very likely, as the current breaker is 90A, but a detail to be checked.
Not sure how to get this information. The label on the AC units are torn. I took pictures of the breakers installed next to the AC units. They are 40A and 30A.
2) The rendering of the panel downstream of the Powerwall isn't accurate. It wouldn't have two 100A breakers in parallel as your diagram currently suggests. (The way your main panel does that is not common.) It would have a new 200A main breaker (in the main breaker position, rather than the backfed 100A branch breaker you currently have) along with the upsized feeder from the main panel to the Gateway to the new 200A main breaker. Which should work as that panel is supposed to have a 200A main bus already. All the existing relocated breakers on its distribution bus would remain unchanged. And a 100A breaker would be used on the distribution bus to supply the panel in the garage. Which, since the existing backfed 100A branch breaker won't be used anymore, could be the exact same breaker in the exact same location, just changing what wires go it.
Got it. Changed the picture to reflect your comment.
Cheers, Wayne
I'm not able to get an answer from Tesla, so hoping you can answer it for me.

In the load calculation that Tesla guys used (NEC 220.83) shows the "Adjusted total of non air conditioning load" as 15312VA and total proposed load current of 63.80A. I understand that the two PowerWalls combined can only deliver 10KW and 60A sustained backup power.

I don't understand why they claim, the sub-panel in the garage can only be supported if I add a third power wall.

Are the load calculations meant to ensure the electrical circuits meet the code and is protected with breakers to prevent overloads? Or, are they borne out of the capacity limitation of the PowerWalls? If it is the capacity limitation of the PowerWall, then the way currently they have wired the PowerWalls, the Oven and the other load will easily exceed the 10KW load!

If the code requirements are met, what is the harm in connecting all the non AC loads to the PowerWalls, even though it might exceed the PowerWalls capacity limit? During an outage, wouldn't the PowerWall protect itself against overload and shutdown when presented with loads that exceeds its capacity?

My reasoning here is that, if everything except the ACs are backed up, it will be incumbent on the user to turn off the excessive loads during an outage and leave it to them to power on what they consider most essential loads.

Wayne, I can't thank you enough for taking the time to answer my questions.
 

Attachments

  • Meter-and-MainPanel.png
    Meter-and-MainPanel.png
    2.6 MB · Views: 13
  • Homeline-Load-center-for-ACs-InsideLabel.png
    Homeline-Load-center-for-ACs-InsideLabel.png
    3.4 MB · Views: 13
  • Homeline-Load-Center-for-ACs.png
    Homeline-Load-Center-for-ACs.png
    3.5 MB · Views: 9
  • ACs-connected-to-HomeLine-Load-Center.png
    ACs-connected-to-HomeLine-Load-Center.png
    4.3 MB · Views: 10
  • OneWire-diagram-proposal.071221.png
    OneWire-diagram-proposal.071221.png
    113.5 KB · Views: 12
  • Tesla-NEC-220.83-Load-Calculations.png
    Tesla-NEC-220.83-Load-Calculations.png
    329.7 KB · Views: 15

wwhitney

Member
Nov 2, 2017
968
1,317
Berkeley, CA
Updated the one line diagram to reflect your comments. Please take a look at the attached picture.
BTW, in my case, the meter and the main panel are adjacent to each other and not on the same panel.
I've attached the pictures for my installation.
If you look at your meter/double main/single distribution, it's actually all one piece of equipment. You can see that the 3R rain hood on top extends across both the left and right sections. So typically that piece of equipment would be a single box on your diagram. (Very minor point.)

I would suggest that the arrows from the new 200A main panel breaker to the Gateway, and from the Gateway to the backed up loads panel (which arrow should point at the its 200A main breaker, perhaps) be made orange and be labeled "new 200A feeder" to make the scope of work clear. Also the orange arrow going to the garage subpanel could be labeled "relocated termination of existing feeder".

On the A/C feeder question of whether the new 90A disconnect is required, it comes down to the existing wire size from the main panel to the A/C subpanel. If it's #4 Cu, that has a 75C ampacity of 85A, so 90A is the maximum allowed protection; if it is #3 Cu, then the ampacity is 100A, and you can omit the 90A disconnect. [For Al, that would #2 vs #1.] Or if the A/C subpanel is not far away, replacing the conductors might be easier than adding the disconnect. [Or it may be possible to use the existing wire along with the tap rules, but for the tap rules the A/C subpanel would need to have a main breaker, which it does not. And adding one would require replacing that subpanel, which would be more trouble than adding a 90A disconnect.] Anyway, you could label the new 90A disconnect with "if required based on existing wire size" or something like that.

I don't understand why they claim, the sub-panel in the garage can only be supported if I add a third power wall.
Skipping for the moment your load calculation questions, my take was that they were saying you would need a second Gateway, because the loads to be backed up are currently split between the two main breakers in your main panel. And as the loads on the current Gateway require 2 PWs, you'd need a 3rd PW for the second Gateway. My proposal avoids the need for the second GW by shifting all the backed up loads to the one 200A main breaker in your main panel.

Cheers, Wayne
 
Last edited:

wwhitney

Member
Nov 2, 2017
968
1,317
Berkeley, CA
On the A/C feeder question of whether the new 90A disconnect is required.
Actually, it's more complicated than my previous answer, as the feeder to the A/C subpanel is primarily for motor loads (plus the 15A breaker for the service receptacle), so the conductor vs OCPD sizing rules are different from typical.. My prior answer was based on the typical rules, but depending on the specs of the A/C units, it might be OK to have a 90A breaker with #8 or #6 Cu conductors.

But if the A/C unit labels are missing, and you can't determine the model numbers, then that information is lost, and you should just assume the 90A disconnect will be required. The information you would need from the labels are MCA (minimum circuit ampacity) and MOCP (maximum overcurrent protection, which will be larger than MCA).

Cheers, Wayne

P.S. What information is on the upper white labels on the A/C units, the labels above the electrocution/hand labels?
 

valleydude

Member
Dec 21, 2019
24
3
California
Actually, it's more complicated than my previous answer, as the feeder to the A/C subpanel is primarily for motor loads (plus the 15A breaker for the service receptacle), so the conductor vs OCPD sizing rules are different from typical.. My prior answer was based on the typical rules, but depending on the specs of the A/C units, it might be OK to have a 90A breaker with #8 or #6 Cu conductors.

But if the A/C unit labels are missing, and you can't determine the model numbers, then that information is lost, and you should just assume the 90A disconnect will be required. The information you would need from the labels are MCA (minimum circuit ampacity) and MOCP (maximum overcurrent protection, which will be larger than MCA).

Cheers, Wayne

P.S. What information is on the upper white labels on the A/C units, the labels above the electrocution/hand labels?
@wwhitney , I updated the 1 wire diagram with your latest comments. The new picture is attached.
I checked the AC units again, they do have the MCA and MOCP listed on the white label:

Unit 1 : MCA - 25.9A , MOCP - 45A
Unit 2: MCA - 19.1A, MOCP - 30A

I've tried to summarize the proposal for rewiring the PowerWalls and will reach out to Tesla and see if they are willing to do it. Please let me know these accurately capture the intent in the 1 wire diagram:
  1. Disconnect the sub-panel in the garage from the 100A disconnect in the main panel
  2. Install a “90A Disconnect” and install a new “100A feeder” from the “100A disconnect” in the main panel
    • The “90A disconnect” may not be required depending on the amapacity of the existing wiring to the ACs.
  3. Install a 4P 200A branch breaker in the main panel
  4. Rewire a new 200A feeder to the Tesla Gateway
  5. Install a 200A main breaker in the “Backup load center”
  6. Rewire a new 200A feeder from the Tesla Gateway to the “backup load center”
  7. Relocate the EVSE breaker from the “main panel” to the “backup load center”
  8. Rewire the existing sub-panel in the garage to the 100A breaker in the backup load center
thanks
 

Attachments

  • Tesl-PowerWall-Installation-rewire-proposal-071521.png
    Tesl-PowerWall-Installation-rewire-proposal-071521.png
    170.9 KB · Views: 8

wwhitney

Member
Nov 2, 2017
968
1,317
Berkeley, CA
Yes, looks good, and the list is correct.

I've researched a bit more the rules for sizing feeders for multiple motors or HVAC units. That makes me think it's very unlikely that you could omit the disconnect, and I'm slightly dubious that the 90A OCPD is the correct size (depends on what size wires are run between the main panel and the HVAC subpanel).

So I suggest changing the diagram to say "New HVAC disconnect (verify existing conductor and OCPD size)" Then FYI,

If the current conductors are #3 Cu, the disconnect could be be omitted (unlikely).
If the current conductors are #4 Cu, then 90A is OK
If the current conductors are #6 Cu, they are large enough for the load in the HVAC subpanel, but the 90A is slightly too large, I think. Calculating the correct size would require close up pictures of both of the labels where you found the MOCP and MCA information; further label data is required for the case of more than one motor or piece of HVAC equipment on a feeder.

Cheers, Wayne
 
  • Informative
Reactions: BGbreeder

valleydude

Member
Dec 21, 2019
24
3
California
Yes, looks good, and the list is correct.

I've researched a bit more the rules for sizing feeders for multiple motors or HVAC units. That makes me think it's very unlikely that you could omit the disconnect, and I'm slightly dubious that the 90A OCPD is the correct size (depends on what size wires are run between the main panel and the HVAC subpanel).

So I suggest changing the diagram to say "New HVAC disconnect (verify existing conductor and OCPD size)" Then FYI,

If the current conductors are #3 Cu, the disconnect could be be omitted (unlikely).
If the current conductors are #4 Cu, then 90A is OK
If the current conductors are #6 Cu, they are large enough for the load in the HVAC subpanel, but the 90A is slightly too large, I think. Calculating the correct size would require close up pictures of both of the labels where you found the MOCP and MCA information; further label data is required for the case of more than one motor or piece of HVAC equipment on a feeder.

Cheers, Wayne
Only one of the AC units label is complete. The other was torn, but, fortunately, I could read the Minimum Circuit Ampacity and "Maximum Fuse or Circuit Breaker" values on the label for both ACs. No other HVAC equipment is connected to the HVAC panel outside the house. Do you think the blower fans in the attic would also be connected to the 90A breaker on the main panel ?
 

Attachments

  • Downstairs-AC.jpg
    Downstairs-AC.jpg
    450.6 KB · Views: 6
  • UpstairsAC.jpg
    UpstairsAC.jpg
    382.1 KB · Views: 5
  • Tesl-PowerWall-Installation-rewire-proposal-071521.png
    Tesl-PowerWall-Installation-rewire-proposal-071521.png
    172.9 KB · Views: 9

wwhitney

Member
Nov 2, 2017
968
1,317
Berkeley, CA
OK, so for the simple case where an HVAC unit has just the compressor and a fan and no other loads, then the baseline data to be extracted from the label is the RLA (rated load amps) of the compressor (or instead, branch circuit selection current, if the label lists it, which should be a bit higher than RLA) and the FLA (full load amps) of the fan. Then MCA = 1.25 * RLA + FLA. And MOCP is typically 2.25 * RLA + FLA, rounded down to the closest standard OCP size (see NEC 240.6(A)). These are supposed to be computed for you, as they are on the label. [If the MOCP is lower than the above, it's because manufacturers sometimes use 1.75 * RLA + FLA.]

For the feeder, the two computations use the 1.25 or 2.25 factor only for the largest compressor, not for both of them. And you add in any other loads on the feeder (in this case, the 15A breaker for the service receptacle; assigning that a load of 15A is conservative, not sure if it's required to be that high).

So if I stick a F in front for feeder, in this example the FMCA = 1.25 * 19.2 + 1.9 + 14.4 + 1.0 + 15 = 56.3. Consulting NEC table 310.16, assuming 75C rated terminals (needs verification), #8 Cu has an ampacity of 50A, too small; #6 Cu has an ampacity of 65A, so that's the smallest allowed size. This also assumes no ampacity adjustment or correction is required (for more than 3 current carrying conductors in one cable or conduit, or for ambient temperature above 30C).

But the FMOCP computation is based on FMCA + 19.2 (the larger RLA, as 2.25 - 1.25 = 1), or 75.5A, rounded down to the standard breaker size of 70A.

The upshot is that the feeder conductors could be #6 Cu (which is most likely, as that is all that is required, and is still the result you get if you incorrectly size based on 15A + MCA1 + MCA2), but then the OCPD should only be 70A, not 90A (which is what you get if you incorrectly size based on 15A + MOCP1 + MOCP2). Or they could be #4 Cu, in which case 90A is correct (the 75C ampacity of #4 Cu). Or if they are #3 or you/Tesla are willing to repull them as #3, then you could use 100A and skip the disconnect.

Cheers, Wayne

PS If you ever see a #14 Cu conductor on a 30A breaker, that can be an allowed configuration for motors or HVAC equipment, per the above sizing algorithm.
 

valleydude

Member
Dec 21, 2019
24
3
California
OK, so for the simple case where an HVAC unit has just the compressor and a fan and no other loads, then the baseline data to be extracted from the label is the RLA (rated load amps) of the compressor (or instead, branch circuit selection current, if the label lists it, which should be a bit higher than RLA) and the FLA (full load amps) of the fan. Then MCA = 1.25 * RLA + FLA. And MOCP is typically 2.25 * RLA + FLA, rounded down to the closest standard OCP size (see NEC 240.6(A)). These are supposed to be computed for you, as they are on the label. [If the MOCP is lower than the above, it's because manufacturers sometimes use 1.75 * RLA + FLA.]

For the feeder, the two computations use the 1.25 or 2.25 factor only for the largest compressor, not for both of them. And you add in any other loads on the feeder (in this case, the 15A breaker for the service receptacle; assigning that a load of 15A is conservative, not sure if it's required to be that high).

So if I stick a F in front for feeder, in this example the FMCA = 1.25 * 19.2 + 1.9 + 14.4 + 1.0 + 15 = 56.3. Consulting NEC table 310.16, assuming 75C rated terminals (needs verification), #8 Cu has an ampacity of 50A, too small; #6 Cu has an ampacity of 65A, so that's the smallest allowed size. This also assumes no ampacity adjustment or correction is required (for more than 3 current carrying conductors in one cable or conduit, or for ambient temperature above 30C).

But the FMOCP computation is based on FMCA + 19.2 (the larger RLA, as 2.25 - 1.25 = 1), or 75.5A, rounded down to the standard breaker size of 70A.

The upshot is that the feeder conductors could be #6 Cu (which is most likely, as that is all that is required, and is still the result you get if you incorrectly size based on 15A + MCA1 + MCA2), but then the OCPD should only be 70A, not 90A (which is what you get if you incorrectly size based on 15A + MOCP1 + MOCP2). Or they could be #4 Cu, in which case 90A is correct (the 75C ampacity of #4 Cu). Or if they are #3 or you/Tesla are willing to repull them as #3, then you could use 100A and skip the disconnect.

Cheers, Wayne

PS If you ever see a #14 Cu conductor on a 30A breaker, that can be an allowed configuration for motors or HVAC equipment, per the above sizing algorithm.
This is awesome information. Thanks for the detailed explanation. Now, how can I determine what type of feeder conductors were used in case? I guess only licensed electrican can tell that, by opening up the panel and looking at the conductors, right?

One other piece of information that I did not include so far is that the AC units are located on the other side of the house, which is almost 70 - 100 feet away from the main panel. Given that, I don't know if Tesla will be willing to repull them as #3, just to avoid installing a new HVAC disconnect.
 

wwhitney

Member
Nov 2, 2017
968
1,317
Berkeley, CA
Well, if you are sufficiently qualified and mechanically inclined, you could crank down the temperature on both A/C units so the condensers turn on, turn off the 90A breaker in your main panel, then go to the HVAC panel, confirm that the condensers have shut off, confirm that the receptacle below the panel no longer has power, and then carefully remove the dead front on the panel. A close examination of the writing on the wires landed on the main lugs will hopefully reveal lettering that says AWG X. Do not break the plane of the front edge of the cabinet with your body or anything conductive, as the above steps to confirm power is off are not complete. (Although I expect many residential electricians would consider them sufficient.)

Cheers, Wayne
 

valleydude

Member
Dec 21, 2019
24
3
California
Well, if you are sufficiently qualified and mechanically inclined, you could crank down the temperature on both A/C units so the condensers turn on, turn off the 90A breaker in your main panel, then go to the HVAC panel, confirm that the condensers have shut off, confirm that the receptacle below the panel no longer has power, and then carefully remove the dead front on the panel. A close examination of the writing on the wires landed on the main lugs will hopefully reveal lettering that says AWG X. Do not break the plane of the front edge of the cabinet with your body or anything conductive, as the above steps to confirm power is off are not complete. (Although I expect many residential electricians would consider them sufficient.)

Cheers, Wayne
I am handy and somewhat familiar with electrical wiring and stuff - enough to make sure I don't electrocute myself :)
 

valleydude

Member
Dec 21, 2019
24
3
California
I am handy and somewhat familiar with electrical wiring and stuff - enough to make sure I don't electrocute myself :)
I opened up the HVAC sub-panel and took some pictures of the conductors. Unfortunately, I could see the markings on the sleeve of the conductor coming from the wall into the sub-panel. Only one of the wires had a marking that is not barely readable. I've attached the pictures.
 

Attachments

  • HVAC-subpanel-without-faceplate.jpg
    HVAC-subpanel-without-faceplate.jpg
    546.1 KB · Views: 8
  • HVAC-conductor-markings.jpg
    HVAC-conductor-markings.jpg
    346.2 KB · Views: 8

wwhitney

Member
Nov 2, 2017
968
1,317
Berkeley, CA
Those are aluminum conductors, and I'm 90% sure that's SER cable. I take it the direct path from the main panel to this panel is entirely through the building?

Looks like you disconnected one of the wires. Now you should get yourself a torque screwdriver, look up the torque rating for the lug that it goes to on that white label on the left of the panel, and tighten the lug to the specified torque to remake the connection. Tighter is not always better.

If you have calipers (preferably non-metallic), you can measure the diameter of both the aluminum conductor (on the stripped end) and the insulation (probably XHHW, I think) and likely determine the conductor size. Since the ampacity of aluminum differs from copper, my previous breakdown doesn't apply. Instead:

#6 Al is too small (75C ampacity is 50A)
#4 Al has sufficient ampacity (65A) but would need to be protected at 70A
#3 Al has a 75A ampacity and may be protected at 80A
#2 Al has a 90A ampacity and may be protected at 90A
#1 Al has a 100A ampacity and may be protected at 100A.

Cheers, Wayne
 

valleydude

Member
Dec 21, 2019
24
3
California
Those are aluminum conductors, and I'm 90% sure that's SER cable. I take it the direct path from the main panel to this panel is entirely through the building?

Looks like you disconnected one of the wires. Now you should get yourself a torque screwdriver, look up the torque rating for the lug that it goes to on that white label on the left of the panel, and tighten the lug to the specified torque to remake the connection. Tighter is not always better.

If you have calipers (preferably non-metallic), you can measure the diameter of both the aluminum conductor (on the stripped end) and the insulation (probably XHHW, I think) and likely determine the conductor size. Since the ampacity of aluminum differs from copper, my previous breakdown doesn't apply. Instead:

#6 Al is too small (75C ampacity is 50A)
#4 Al has sufficient ampacity (65A) but would need to be protected at 70A
#3 Al has a 75A ampacity and may be protected at 80A
#2 Al has a 90A ampacity and may be protected at 90A
#1 Al has a 100A ampacity and may be protected at 100A.

Cheers, Wayne
@wwhitney Thanks for patiently listening to me and providing me with a possible solution.
I wish Tesla can learn something from you. I've reached out Tesla Energy and shared your rewire proposal with them.
Waiting to hear back from them.

About the torquing the lug nut, thanks for pointing that out. I actually tightened it. Did not realize the electrical screws also need to be torqued appropriately. I will check the torque rating for the lug and tighten it to the spec.

The direct path from the main panel to the HVAC subpanel through the house.
 

Products we're discussing on TMC...

About Us

Formed in 2006, Tesla Motors Club (TMC) was the first independent online Tesla community. Today it remains the largest and most dynamic community of Tesla enthusiasts. Learn more.

Do you value your experience at TMC? Consider becoming a Supporting Member of Tesla Motors Club. As a thank you for your contribution, you'll get nearly no ads in the Community and Groups sections. Additional perks are available depending on the level of contribution. Please visit the Account Upgrades page for more details.


SUPPORT TMC
Top