2

It seems that the default for electric storage-type water heaters is two 4500watt heating elements, wired for non-simultaneous operation.

I'm curious why that is. Water heaters are typically on 30amp breakers, and 30amps * 240volts * 80% = 5760 watts (80% being the de-rating factor for a circuit's ampacity for "continuous" loads).

Put another way: is there some reason I shouldn't be able to replace the 4500 watt elements in my water heater with 5500 watt ones ? Bradford-White, for one, allows one to order water heaters with elements in 500 watt steps from 1500 through 6000 watts (so presumably they can supply heater elements in these wattages).

3

My belief is the controls are cheaper. I have installed a 5500w element in my home’s water heater after the element had failed. The control was rated for 4100 or 4500 I don’t remember, and within 6 months that control or thermostat went out. It was a high end model, so I replaced the controller and it only lasted 4 months. I called for a warranty (providing my license info) and they sent a new one. The new one also failed in 4 months, so I went through all the fun stuff again and the company said for $20 more I could upgrade to the commercial one. I told them they could send me that one on their dime; I said it was their warranty and was listed for the element they tried to convince me to pay the extra saying 4500watts is normal and I had to upgrade. Short version: they did send the commercial version after verification of my license (where they get home owners). I had disassembled the first one that failed the contact sets were fairly small about the diameter of a pencil maybe slightly larger. The commercial version the contacts were replaceable and close to 1/2” square. That water heater was still running years later when I sold that house so I believe the reason most come with smaller heaters is they are on the edge of needing larger contacts and this costs more. My example is from the late 80’s but contact amperage would be a constant increase the load and a larger area in needed for long life.

The company tried to get out of the 3rd replacement by saying the instructions required a licensed electrician I faxed my card to them, that’s when I got the commercial replacement. It had a higher rating but I no longer remember what it was.

So I believe that they provide lower wattage elements for longer life. When the controller failed they wanted proof it was installed by a licensed electrician. I bet they do it on the first failure today.

7
  • Oh, so you did the calculation I did (in my OP) and said "dammit, I want more hot water and a 5500 watt element should be safe". Dec 12 '20 at 1:30
  • Sure the higher wattage/current requires slightly beefier contacts; but it would've been a simple matter, back when someone decided what the "standard" wattage should be, to do the calculations we've done here and make the standard be 5500 watts (or 5000 watts, if manufacturing tolerances are as big a factor as suggested below); I doubt such a contact would've been significantly more expensive. I'm honestly surprised they cut it so close that you saw those failures. Dec 12 '20 at 1:33
  • 1
    I wonder, if one special-ordered higher-wattage elements, if they'd put beefier thermostats in ? I DID special order a 3000watt element in the 10 gallon unit I put in our kitchen, and it's actually simple enough and did not even cost me more. Dec 12 '20 at 1:35
  • 1
    We used to replace the contacts in contactors (big relays) but about 20 years ago the contact kits became as expensive as a complete replacement, not sure why, the manufacturing tolerance of even 20% is nothing because the tolerance is based on resistance not wattage sorry internet only electricians. The voltage/ resistance = amps so we don’t even have to get fancy to figure that out.
    – Ed Beal
    Dec 12 '20 at 2:14
  • It's not obvious to me that better than 20% tolerance is ridiculous. At least not because the variable is resistance. Resistors (for circuits) with 1% tolerance are commonplace and cheap, and even for wire-wound ones, 5-10% is common. Dec 12 '20 at 5:16
0

In areas where houses are supplied at 208 volts, the calculation yields 4992 watts. The first 500watt step below this is 4500 watts. Even so, you'd think 4800 watt elements would be fairly common.

9
  • Then factor in some manufacturing tolerance - are all water heating elements the exact same resistance? There is a tolerance like +\- 10%.
    – Solar Mike
    Dec 11 '20 at 8:07
  • 1
    The heater doesn't automatically draw 80% of its breaker rating - it draws V/R amps. If it's rated 4500 W (24 A) at 240 V then it's a 10-ohm resistor. The actual heat delivered when powered at 208 will be less (V^2/R=4326 W).
    – Greg Hill
    Dec 11 '20 at 16:35
  • @GregHill Check your math on 24x240. Dec 11 '20 at 22:29
  • 1
    @Harper-ReinstateMonica Awww come on, it's Friday. ;-) Thanks. So, as I so poorly expressed by using numbers recklessly: if a heater's rating of 4500 W is assigned with expectation it'll be fed at 240 V it would draw 18.75 amps from that voltage (12.8 ohms). If it were supplied with 208 V, though, it would draw only 16.25 amps and thus yield 3380 W of heat. Point is, it isn't clear what 80% of a 30A breaker on a 208 volt system has to do with 4500 W being a common heating element.
    – Greg Hill
    Dec 12 '20 at 0:01
  • Good point, guys. I assumed my proposed "5500 watts at 240 vac" element would still produce 5500 watts at 208v, and thus would draw more current and exceed 80% of a 30amp breaker - when in fact it would draw less current, and not exceed 80% * 30amps. Dec 12 '20 at 1:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.