Current Leakage Between Heater and Kettle?

Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum

Help Support Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Skeeter686

Supporting Member
HBT Supporter
Joined
Sep 19, 2022
Messages
169
Reaction score
267
Location
Illinois
I just had my brew day disrupted by my GFCIs tripping (both on the controller and the wall outlet). After reverting to propane, I broke out a multimeter and checked some connections.

The cord connecting my controller to the heating element seems to be good. I checked between each pin and the resistance is beyond the capabilities of my (cheap @$$) multimeter to measure.

When I checked the heating element, I get 0 resistance between the one pin and the TC clamp, so this must be the ground. For the other two pins, I'm measuring less than 1M-Ohm (somewhere around maybe 700k- to 900k-Ohm) resistance, but there's obviously not the same level of isolation that I see in the cord.

Is it possible that I have a defective heating element and that it's leaking enough current to cause my GFCI to regularly trip?

This is a Spike unit, but I don't yet have a 240V outlet available, so I'm running it with an Inkbird at 120V currently. That's why I'm trying to troubleshoot this, myself, before contacting the manufacturer.
 
-How much current is each of these; -GFCI, -SSR, -outlet/breaker rated for?
-Have you used this unit previously, even for a test run?
 
fwiw, between the NEC and UL 943 there seems to be agreement that Class A GFCI devices "shall" trip with a threshold leakage current between 4 to 6 milliamps. Theoretically your 700K ohm resistance could only pull ~0.2 milliamps (200 microamps), well below that threshold.

However...measuring leakage current with low voltage DC using a cheap ohmmeter may be fraught with error. Also, measuring the leakage of a resistive heating element when it's cold may miss the crux of the issue - that 700-900K may be dropping precipitously?

Cheers!
 
-How much current is each of these; -GFCI, -SSR, -outlet/breaker rated for?
-Have you used this unit previously, even for a test run?
The Inkbird controller is rated for up to 1800W, so that should be 15A. It has an integrated GFCI in the plug, which tripped.

The GFCI wall outlet appears to be a 20A outlet, based on the shape of the slots in the receptable.

The house circuit breaker for that line is 20A.

The heating element for the unit is 5500W at 240VAC, which I believe should be 1375W for the 120VAC that the Inkbird supports. So, it should be well within the limits of the controller, based on my non-professional understanding of electricity.

I have used this unit previously (both for a cleaning test run and for one brew day) and I didn't have this problem. It does strike me as odd that I'm only seeing the issue now. However, the GFCI on the Inkbird always seemed a bit sensitive and it seemed that I'd usually need to reset it after plugging the unit in. I've had experience with overly sensitive GFCI outlets in the past, so I didn't think much of it. Since it would occur even without activating any of the outlets on the controller, it didn't seem to be related to the attached equipment.
 
I looked at your other posts to try and determine your setup and found you have the Inkbird IPB16S...When I frist looked at ready-made controllers years ago I dismissed it as my instinct just balks at the idea of a sustained load to bring >5 gallons to a boil on a 15A circuit, but I did log it as a potential RIMS controller for maintaining an already achieved temp. I've since dismissed it though, opting to build an Auber DSPR320 based unit based both on @Bobby_M 's descriptions on this site, and the reviews of the inkbird on his site: Temp Controller, 120V Plug and Play PID (Inkbird IPB-16S) ...in the description he even states: "It is the best controller for use with a 120 volt RIMS tube." and there is no suggestion of achieving or maintaining over time a large boil.
The unit itself may well have an overly sensitive GFCI, but the occasions where I've read about it doing general controller-research online, I've only found a few people happily using it on thier RIMS tubes, and many people disappointed with it for that as well as any other application. I suspect it simply can't handle the load you're asking of it.
That said, I know many folk do use low power systems (and hopefully some of them will weigh in here), but their accounts of it always seem to include slowness and/or home-built or more expensive controllers.
 
I looked at your other posts to try and determine your setup and found you have the Inkbird IPB16S...When I frist looked at ready-made controllers years ago I dismissed it as my instinct just balks at the idea of a sustained load to bring >5 gallons to a boil on a 15A circuit, but I did log it as a potential RIMS controller for maintaining an already achieved temp. I've since dismissed it though, opting to build an Auber DSPR320 based unit based both on @Bobby_M 's descriptions on this site, and the reviews of the inkbird on his site: Temp Controller, 120V Plug and Play PID (Inkbird IPB-16S) ...in the description he even states: "It is the best controller for use with a 120 volt RIMS tube." and there is no suggestion of achieving or maintaining over time a large boil.
The unit itself may well have an overly sensitive GFCI, but the occasions where I've read about it doing general controller-research online, I've only found a few people happily using it on thier RIMS tubes, and many people disappointed with it for that as well as any other application. I suspect it simply can't handle the load you're asking of it.
That said, I know many folk do use low power systems (and hopefully some of them will weigh in here), but their accounts of it always seem to include slowness and/or home-built or more expensive controllers.
I'm planning to start getting quotes for installing a 240V outlet, but I expected that I could use this at 120V to maintain a stable mash temperature. I'm not planning to use it for the boil (I'll transfer to my old boil kettle and use my propane burner for now).

I do know someone who claims that 120V is sufficient for a boil, but even he states that it's not a very vigorous boil. Considering how long it would take to even get to boil, I didn't want to bother with that.

The thing that concerns me is that my outlet also tripped, which means that it's most likely not just an overly sensitive GFCI on the Inkbird.

I'll also try plugging in my Kill-a-watt to see how much is actually being drawn when in use.
 
I'll also try plugging in my Kill-a-watt to see how much is actually being drawn when in use.
Good plan! As to the getting qoutes..you may be able to delay it a bit depending on where you brew in your house and wether or not you have a 240V stove or dryer outlet. I brew in my kitchen by rolling in my brew cart, sliding the stove away from the wall and using an adaptor-plug I made. [As to the dryer, I'd moved by accident of circumstance to a 120V unit, so I pulled, rerouted and repurposed the supply line to a backyard 240V 30A outlet for my MIG welder and plasma-cutter...but I almost put my brew-rig in the laundry room to share the plug]
 
Good plan! As to the getting qoutes..you may be able to delay it a bit depending on where you brew in your house and wether or not you have a 240V stove or dryer outlet. I brew in my kitchen by rolling in my brew cart, sliding the stove away from the wall and using an adaptor-plug I made. [As to the dryer, I'd moved by accident of circumstance to a 120V unit, so I pulled, rerouted and repurposed the supply line to a backyard 240V 30A outlet for my MIG welder and plasma-cutter...but I almost put my brew-rig in the laundry room to share the plug]
Unfortunately, I've got gas. And does does my stove and dryer. There's no 240V outlet in the house at present.

I tried to check the heating element with the Kill-a-watt, but the GFCI trips instantly.

If I unplug the cord (that I made) from the heating element, nothing happens when I turn on the heat, which reinforces that the problem is not with my cord. If the plug is connected to the heating element, the GFCI trips instantly when the heating circuit activates.

I can successfully run the pump from the other outlet on the Inkbird (it draws between .8A and .95A, according to the Kill-a-watt).

I might see if I can find someone with a 240V outlet so that I can test this with the Spike controller. My BIL has a welder, so I'm guessing he's got an outlet I could use. I have a feeling I'll be calling for a replacement heating element, though.
 
I didn't read every word in the thread so I apologize if you tried this already. Plug your home made cord WITHOUT the element connected, into just the wall outlet (that I believe you said is GFCI also). Does that trip? Do it again with the element on the cord. Trip?
 
I didn't read every word in the thread so I apologize if you tried this already. Plug your home made cord WITHOUT the element connected, into just the wall outlet (that I believe you said is GFCI also). Does that trip? Do it again with the element on the cord. Trip?
Thanks for the idea. The cord alone does not trip the GFCI when connected directly to the wall outlet. If the heating element is connected to the cord, then the GFCI trips instantly.

At this point, it seems obvious that there's an issue with the heating element.
 
Any chance there's water in the element enclosure (if you have an enclosed element, that is)?
I was wondering about that, myself. I didn't notice any obvious defects in the element "sheath" but maybe it's not completely sealed in the section where it passes through the TC fitting.
 
Any minor or major short between the resistive wire loop and the metal sheath will cause this. 99% of the time this happens, it's due to a dry fire that causes the element to get so hot that it changes shape. Checking with an OHM meter the resistance between the two hot prongs should be about 9-10ohms. Between either hot and the ground prong, the reading should be infinity. If it's anything lower than that, you've got a short.

Add the cord to the element and do the same measurements on the 120v male end of the cord. Same resistance values should be measured there.

The one thing that would cause tripping for sure is if you mistakenly have the ground/neutral mixed up on your 15 amp 120v plug end. That would send current down the ground instead of the neutral and GFCI is specifically looking for that fault. You can easily check your DIY cable by disconnecting it from everything and setting the meter to ohms. Put a probe in the L6-30R slot that has the hook on it and then the other end touched to the round plug (ground). That should ready zero ohms.
 
Any minor or major short between the resistive wire loop and the metal sheath will cause this. 99% of the time this happens, it's due to a dry fire that causes the element to get so hot that it changes shape.
I'm sure that you nailed it. I came to this conclusion last night, myself.

I'm new to electric brewing. It wasn't until later last night that I realized that the element was much darker than after the first time I'd used it. Plus, I was distracted by my concern that my modification was possibly the issue.

What I believe happened was that I was mashing a relatively small batch and I probably had the pump running too aggressively to circulate the liquid back to the top of the grains. It was probably drawing liquid faster than it drained back through the grains, creating a void. Once the element was no longer submerged, it overheated and was damaged, causing the the current leakage. Then the GFCI did was it was supposed to do and it killed the circuit.

The liquid eventually refilled the bottom of the kettle so that I didn't notice the issue by that point. Also, being new to electric brewing, the obvious overheating of the element didn't immediately jump out at me.

Lesson learned. I think that I'll probably split the flow so that I can get good circulation via the whirlpool port and have more of a trickle flowing down from the top.

This wasn't something that I think I'd easily notice during operation because you can't really see the liquid level under the grains.

Oops!
 
This makes me wonder: analogous to the Auber boilover prevention capability, could a controller use a temperature sensor or, perhaps, a current sensor, to quickly detect and suppress dry firing?

Current sensing could be irrelevant if dry firing uses a normal amount of current. And sensing the temperature of the element could pose difficulty.

It could be preferable to engineer protection into the element itself. My only dry fire event, IIRC, seemed to temporarily disable the element, which has since worked fine.

So this speculation could be pointless...
 
This makes me wonder: analogous to the Auber boilover prevention capability, could a controller use a temperature sensor or, perhaps, a current sensor, to quickly detect and suppress dry firing?

Current sensing could be irrelevant if dry firing uses a normal amount of current. And sensing the temperature of the element could pose difficulty.

It could be preferable to engineer protection into the element itself. My only dry fire event, IIRC, seemed to temporarily disable the element, which has since worked fine.

So this speculation could be pointless...
Might be possible to monitor the resistance of the element in real time (less than 1 second response time in this case.) Most resistors change resistance a little with temp changes (that's how an RTD works after all.) The bigger problem is that you most likely start with a partial dry fire, where the element is partially submerged and partially exposed to the air. Resistance monitoring might not be able to detect this condition.

Brew on :mug:
 
So, the GFCI plugs sense if you have any leakage from hot to the ground. The amount of load shouldn't be part of the picture.

Leakage like a human getting across the circuit, and that's pretty small before it gets harmful, so it's a pretty high resistance/pretty low current threshold. It would be in the 100K of ohms, don't know off the top of my head what the trip current is.

For just the element doing this, it's the insulation inside, between the actual heating element and the metal outer tube that protects it from water and or being a shock hazard. The theories of overheating causing warping and damage, are pretty plausable, and probably how you got there.
Either the element bending and nearly pushing thru the internal insulation, or maybe a tiny crack in the outer tube, letting moisture in... either way, that would do it. And, either way, a new heater element is what you're looking at.

Interesting idea, on how to detect a dry heater element... probably kinda challenging. I'd guess that the resistance tempco would be pretty small in these, and hard to measure. Otherwise the power they would use would go up and down with temp too much. The kinds of stuff they usually make them of has a tempco, but it's not noticiable till you get up 1K degrees or more... Maybe you'd have to build it with a thermocouple inside, then have to have an external monitor/alarm/cutoff circuit... A lot of added complexity and $$...
 
I think everyone that moves to a recirculating electric BIAB system eventually learns the recirculation rate limit the hard way. It's one of things that you only have to learn once.
I guess it's nice to know that I'm not the only one. It is a rather costly and frustrating mistake... definitely something I only need to learn once.
 
Restrict your recirculation rate to 1-2 quarts per minute to reduce the possibility of repeats. The only absolute way to prevent it, especially if you walk away for a while regularly, is to put a float switch in.
 
I was also wondering how to detect and prevent this in the future. The heating element is at a certain level in the kettle, so monitoring the level of liquid to ensure it doesn't drop to (or below) the level of the element seems like a good way to go.

For manual monitoring, a sight glass aligned at a certain level could let you visually check. But much of the value of a system like this is to use some automation to free up the brewer's time.

Not sure what the best approach would be for a water level sensor that would reliably hold up to the pH, minerals, and temperatures over the long haul. But that seems like a better option than monitoring the element. By the time you would see a change in the current drawn by the element, I suspect it would already be eating itself and it would be too late.
 
I was also wondering how to detect and prevent this in the future. The heating element is at a certain level in the kettle, so monitoring the level of liquid to ensure it doesn't drop to (or below) the level of the element seems like a good way to go.

For manual monitoring, a sight glass aligned at a certain level could let you visually check. But much of the value of a system like this is to use some automation to free up the brewer's time.

Not sure what the best approach would be for a water level sensor that would reliably hold up to the pH, minerals, and temperatures over the long haul. But that seems like a better option than monitoring the element. By the time you would see a change in the current drawn by the element, I suspect it would already be eating itself and it would be too late.

The one approach some Spike Solo users have been using is drawing from the whirlpool port with the bent tube rotated upward so that if the liquid level was depleted, the pump would suck air and make a noise you could recognize from within ear shot. I'm not crazy about using a partial run-dry situation as an alarm. There are stainless steel magnetic float switches that can handle the rigors of brewing. The easiest way to implement it is to run the SSR 12VDC control signal through the switch. If the level is low, the element won't fire. When level is replenished, it will.
 
I was also wondering how to detect and prevent this in the future.
Direct firing the mash can bring many potential issues, such as yours.

To solve this, a brew friend of mine, using a 20 gallon system, built an "Indirectly Heated RIMS" for (indirectly) heating the recirculating wort.

In essence, a relatively small, 6' stainless coil for recirculating wort from the mash, is placed inside a (smallish) 1 gallon pot with water. That water in the pot is directly heated with a cheap water heater element switched on by a PID controlling the mash temp.

It works very well, without scorching.
IIRC, he also included a secondary control to prevent the wort in the RIMS coil getting too hot, so enzymes aren't denatured prematurely, until the mashout step.
 
Direct firing the mash can bring many potential issues, such as yours.

To solve this, a brew friend of mine, using a 20 gallon system, built an "Indirectly Heated RIMS" for (indirectly) heating the recirculating wort.

In essence, a relatively small, 6' stainless coil for recirculating wort from the mash, is placed inside a 1 gallon pot with water. That water is directly heated with a cheap water heater element switched on by a PID controlling the mash temp.

It works very well, without scorching.
IIRC, he also included a secondary control to prevent the wort in the RIMS coil getting too hot, so enzymes aren't denatured prematurely, until the mashout step.
Yes, I was concerned about scorching the wort and was very interested in an approach that a friend of mine cobbled together (vaguely similar concept to what you describe). But I decided to think long term and there are a lot of people having success with all-in-ones, so I decided to go with the Spike Solo as a convenience and to hopefully let me brew more often.

I guess I didn't spend enough time reading about them, because I really should have caught my mistake before it happened. I'm not usually so dense! 🙄
 
... or "mini-HERMS" I suppose. Cool (hot) idea.
A friend of mine runs a system where he submerges a wort chiller in the mash and uses an Inkbird to circulate hot water through it in order to maintain mash temperature. He heats the water on a burner and only needs to maintain a fairly wide temperature range in order for the system to keep a consistent mash temperature. It works pretty well for him and definitely won't scorch the wort.
 
Back
Top