Radiant Heat Fails Megohm Pretest

[mkgrady]

A friend of mine is having under floor electric radiant heat installed by a tile installer. The installer layed out the wire on the floor. My friend asked for the results of the Megohm test. The value must be at least 1000 megohms at a test voltage of 1,000. The tile installer responded that he just uses a multi meter to confirm the ohm value of the heater. Meaning he does not bother with the Megohm test thereby voiding the warranty. Shame on the tile guy for Allowing this step to be skipped.

Of course the test should be done by an electrician so my friend called me to conduct the test. I performed the test using my extech 380260 Megohm meter. Both heater cables failed the test. One cable read 460 megohms and the other read 160. The tile guy then rips out the newly installed heater cable and heads to his supplier to complain that the cable is defective. He returns with two new cables and they also fail the test.

The next day the tile installer shows up with a borrowed fluke Megohm meter and this time the cables pass the test. Each cable tested out at 2,200 megohms.

The likely conclusion is that my meter is defective but how can that be determined?

 

[GoldDigger]

How? Send it for calibration if you want to be sure.
Compare it on same test wiring against a suspected good unit *at the same voltage setting*!
Do you know what voltage you used and what voltage the third party used, and how both compare to the manufacturer's test specs?
Not all Megger measurements are equal.

 

[kwired]

A friend of mine is having under floor electric radiant heat installed by a tile installer. The installer layed out the wire on the floor. My friend asked for the results of the Megohm test. The value must be at least 1000 megohms at a test voltage of 1,000. The tile installer responded that he just uses a multi meter to confirm the ohm value of the heater. Meaning he does not bother with the Megohm test thereby voiding the warranty. Shame on the tile guy for Allowing this step to be skipped.

Of course the test should be done by an electrician so my friend called me to conduct the test. I performed the test using my extech 380260 Megohm meter. Both heater cables failed the test. One cable read 460 megohms and the other read 160. The tile guy then rips out the newly installed heater cable and heads to his supplier to complain that the cable is defective. He returns with two new cables and they also fail the test.

The next day the tile installer shows up with a borrowed fluke Megohm meter and this time the cables pass the test. Each cable tested out at 2,200 megohms.

The likely conclusion is that my meter is defective but how can that be determined?

Did tile installer test with a 1000 volt test volts?

 

[mkgrady]

How? Send it for calibration if you want to be sure.
Compare it on same test wiring against a suspected good unit *at the same voltage setting*!
Do you know what voltage you used and what voltage the third party used, and how both compare to the manufacturer's test specs?
Not all Megger measurements are equal.

All tests were performed at 1000 volts. I would like to believe my tester is good and theirs is bad but I'm guessing it is the opposite. Can I really get a merger calibrated for less than the cost of a new one. Mine cost about $250. The fluke is about $500

 

[Dennis Alwon]

I think $250 is fairly inexpensive for a megameter.

Seriously I have installed many floor heat systems and the only thing we use is the loud mouth detector that they send with the unit. Something like this

 

[K8MHZ]

All tests were performed at 1000 volts. I would like to believe my tester is good and theirs is bad but I'm guessing it is the opposite. Can I really get a merger calibrated for less than the cost of a new one. Mine cost about $250. The fluke is about $500

If I had to guess, and my choices were Extech vs. Fluke, I would go with Fluke.

That being said, a guess really has no value.

The first step would be to compare your meter to one or two different meters to see if yours is really that far off.

A reading of 2,200 megs sounds suspicious to me. I sounds like the leads weren't even connected.

I would re-test using a third meter. You said the Fluke was borrowed so there really is no way to know if it actually works or not.

 

[ceknight]

If I had to guess, and my choices were Extech vs. Fluke, I would go with Fluke.

That being said, a guess really has no value.

This thread brings back memories...of why I have an Extech megger. A while back my supply house was carrying a brand of floor warming mat, don't remember the name off the top of my head but it was from Canuckistan. And there in the fine print was a requirement to megger it before installation. So I needed a megger in a hurry, and the only place in town that had one in stock that day was Grainger, and it was the Extech. So I didn't have a choice.

But the mat passed the test, and I've had no other reasons to expect a problem with the meter. So for the OP I'd suggest having it tested/calibrated. Even at under $300 it shouldn't be a disposable tool.

 

[meternerd]

This thread brings back memories...of why I have an Extech megger. A while back my supply house was carrying a brand of floor warming mat, don't remember the name off the top of my head but it was from Canuckistan. And there in the fine print was a requirement to megger it before installation. So I needed a megger in a hurry, and the only place in town that had one in stock that day was Grainger, and it was the Extech. So I didn't have a choice.

But the mat passed the test, and I've had no other reasons to expect a problem with the meter. So for the OP I'd suggest having it tested/calibrated. Even at under $300 it shouldn't be a disposable tool.

Meggered line to what? If it's not installed, what do you megger it to? The ground wire (assuming it has one)? Most I've seen fail are due to installation damage. Why so high of a reading needed. At 120V, 3 or 4 hundred Megs is a miniscule amount of leakage current. If it was mine, I'd energize it and measure current. Meggers are really not intended for high accuracy readings. Temp, humidity, etc. can affect the results. Really more of a pass/fail deal. Energizing the cable for a while and then retesting might have really changed the results. Just my experience, though. I like Fluke, by the way.

 

[ceknight]

Meggered line to what? If it's not installed, what do you megger it to? The ground wire (assuming it has one)?

Line to ground. I remember now, it was made by Ouellet.

 

[kwired]

Meggered line to what? If it's not installed, what do you megger it to? The ground wire (assuming it has one)? Most I've seen fail are due to installation damage. Why so high of a reading needed. At 120V, 3 or 4 hundred Megs is a miniscule amount of leakage current. If it was mine, I'd energize it and measure current. Meggers are really not intended for high accuracy readings. Temp, humidity, etc. can affect the results. Really more of a pass/fail deal. Energizing the cable for a while and then retesting might have really changed the results. Just my experience, though. I like Fluke, by the way.

I believe some cables have a metallic shield on them. The few I have seen installed (and were done by the tile guy) they cut grooves in the substrate, lay the cable in the groove and out an epoxy covering over entire substrate and cable. The then test cable before they lay tile over it, should it fail these tests (which may include a megger test) they either repair or replace and test again. I think they typically test again before grouting the tile, just in case there may have been any damage during tile installation.

 

[Sierrasparky]

It also is about knowing your tool. If you use it often you will get a feel for when the tool is giving a bogus result. Use the megger on known wires. or compare with another megger that is known to be good.

You might want to check for calibration.

 

[Smart $]

All tests were performed at 1000 volts. I would like to believe my tester is good and theirs is bad but I'm guessing it is the opposite. Can I really get a merger calibrated for less than the cost of a new one. Mine cost about $250. The fluke is about $500

I have no idea what calibration costs, but if you don't want to go that route, at the very least get say a 10 megohm resistor for comparison. They are quite inexpensive (a matter of pennies), though packaging, marketing, and delivery drive up the cost to purchase just one.

 

[GoldDigger]

Note, however, that 1000V across one megohm is one watt, so do not just grab a 1/4 watt resistor.
Not that you intend to leave the Megger connected very long.

 

[mkgrady]

Does it even make sense that two different cables would give the exact same test result? The fluke also read 2200 megohms with the leads not connected to anything.

 

[Smart $]

Note, however, that 1000V across one megohm is one watt, so do not just grab a 1/4 watt resistor.

That's why I said 10 megohm resistor. At 1000V, only 0.1W. Even an 1/8 watt resistor will work.

 

[GoldDigger]

That's why I said 10 megohm resistor. At 1000V, only 0.1W. Even an 1/8 watt resistor will work.

Lost count of the zeros.

 

[meternerd]

Does it even make sense that two different cables would give the exact same test result? The fluke also read 2200 megohms with the leads not connected to anything.

All digital meggers have a full scale (infinity) value. The Fluke we use is 11G. On yours it's 2.2G. Your are reading infinity.

 

[growler]

The tile installer responded that he just uses a multi meter to confirm the ohm value of the heater.

What ohm value? Ask him if he ever reads the installation instructions because that is the only way to determine the correct test proceedure.

The last time I got involved in floor heating cable I read the instructions and the cable failed. But the cable didn't match the numbers of the installation instructions. So I called the manufacturer to check it out and they wanted to know where the cable was purchased. The tile guy said the owner had purchased cable from Amazon. They said that often the wrong installation manual is shipped and when the correct installation manual/instructions arrived ( via e-mail) the cable passed.

What I'm getting at is there is no end to the number of things that can go wrong.

First make sure you have the correct test values and then have the cable rolled out and test before it can be damaged. If it fails then rent a top of the line megger.

One thousand meg at 1000V sounds a bit much to me but if that's what they want that's what they should get.

 

[PetrosA]

In my experience with a few, limited brands of floor heat cable, megger tests were not required for testing according to instructions. A good DMM is. The cable brand I use most often is Nexans which has a tag with factory readings on each length of cable for resistance. That must be verified before unpacking, after installation, after wet bed/embedding and again before energizing along with a continuity test between heater wires and ground. A megger test is recommended for this step, but not performing it will not void warranty.

I honestly lost trust in Extech tools a few years back when I owned their high-end clamp meter (EX830 IIRC). It was giving readings all over the place on any setting after about one year. There are certain companies that are known for making quality megohmmeters and I would stick with one of them - AEMC, Fluke, Hioki, Megger are the main ones. I currently use a wind up unit from AEMC that I'm very happy with. I love the immediate feedback that you get when there's a short in the line and never having to replace batteries

A note on calibrations: There's a lot of confusion about what a calibration is, and I'm seeing it in this thread as well. A calibration DOES NOT mean that a technician regulates a meter to show accurate readings. It is a PASS/FAIL measure of what the meter readings are at predetermined standard values to indicate how close to the standard the meter IS ABLE to read. Some meters can be readjusted to show good readings for the standard values but the problem with doing this is as follows: If a meter is adjusted and calibrated at the factory and those readings drift drastically over a period of time, there is a problem with the meter itself. Readjusting it to show accurate values will not guarantee accurate readings at a later point in time. If you have a megger with stated 1.5% accuracy that was calibrated at the factory and it read 998 Ohms for the 1000 Ohm standard at that time, and five years later it reads 993 Ohms for the 1000 Ohm standard, you have a very accurate meter with very little drift. If, five years later, it reads anything over 1015 Ohms or below 985 Ohms for the 1000 Ohm standard, that meter is trash. The meter has failed calibration.

Periodic calibrations validate the readings obtained between calibrations. If a meter is calibrated once per year, the last passed calibration indicates that any tests performed between calibrations have been accurate. If a meter fails calibration, any tests performed since the last passed calibration may be questioned, since there is no way to tell when the meter went out of calibration. That's the problem I saw with Extech tools. They would start giving bad readings and there wasn't any way to tell when they started getting wonky.

 

[mkgrady]

What ohm value? Ask him if he ever reads the installation instructions because that is the only way to determine the correct test proceedure.

The last time I got involved in floor heating cable I read the instructions and the cable failed. But the cable didn't match the numbers of the installation instructions. So I called the manufacturer to check it out and they wanted to know where the cable was purchased. The tile guy said the owner had purchased cable from Amazon. They said that often the wrong installation manual is shipped and when the correct installation manual/instructions arrived ( via e-mail) the cable passed.

What I'm getting at is there is no end to the number of things that can go wrong.

First make sure you have the correct test values and then have the cable rolled out and test before it can be damaged. If it fails then rent a top of the line megger.

One thousand meg at 1000V sounds a bit much to me but if that's what they want that's what they should get.

Each heater cable comes with a sticker from the factory that shows what the ohm value should be.