TC beyond Ni200: Nickel Purity, Dicodes; Ti, SS, Resistherm NiFe30; Coefficient of Resistance

[notarobot]

i'm writing an email to them now, would be great to have as many points as possible to include as to why this does not seem to be anything like what we expected.

Nickel is heavy... One meter of 0.4mm SS should have a weight of ~1.00g. For NiFe30 it should be 1.06g.

 

[TheBloke]

OK I'll leave out the magnetism

@druckle - if they have supplied a wire that is 70% Nickel and 30% Iron then they can honestly say they've provided what people have asked for. Indeed, their listing specifically says that it is 70% Nickel and 30% Iron. (Or to be precise, it says it's Ni70Fe30 wire.)

So the question is: could a Ni70/30Fe wire be so very different in TCR, resistance and magnetism to the other Ni70Fe30 wires we've looked at? Maybe it can - certainly we know the TCR can very at least from 0.0032 up to 0.005, maybe that can go all the way down to 0.001. Maybe it can be variably magnetic. Maybe resistance of one Ni70Fe30 can be 4x higher than another Ni70Fe30.

I know trace elements and minor differences in the balance of the metals can vary all these things. I just hadn't realised it could vary quite that much.

If we were talking anyone besides Crazy, I would assume that yes it must be that it can vary that much. But this is Crazy Wire. They have a.. history.

@notarobot ah excellent! Let me check that now.

 

[TheBloke]

@notarobot when you said SS, will any SS do? I have 0.40mm SS304. I have 316 and 317 but not in 0.40

I want to able to weigh the SS as well to confirm the comparison.

 

[TheBloke]

Actually never mind, I just realised my scales only go to one decimal, so unless I weigh 5m of the stuff or something, that tiny difference is going to be hard to check for

I knew I should have got the scales that did 2 decimals.

 

[notarobot]

SS 304/3416/317 density differs by at most 1%. One meter of 0.4mm SS 304 should be 1.01g.

 

[druckle]

This paper may help in evaluating the alloy you received compared to what you were expecting.
It has a pretty good breakdown of the iron nickel alloys and their expected propertiesl

It might be a bit of a slog but it's pretty straight forward.

http://www.asminternational.org/doc...A9D7.pdf/d490dee6-620e-4e38-b64d-53dd02c5fc81

Duane

 

[TheBloke]

For the fun of it I weighed 0.6m of SS304 and the Crazy Mystery Wire and both showed as 0.6g. Doesn't tell us much, just that they're not majorly dfferent.

I may try weighing 3m later, I have 25m of this stuff and it doesn't look like I'm ever going to vape on it. That would make the difference nearly .2g and hopefully detectable.

 

[soulseek]

Guys as always great work, keep it up. Personally, I don't see myself moving away from Ti, even if it is still a bit springy when annealed. Maybe I could see myself trying SS out of curiosity. For now I'll keep enjoying my juice on my DNA200 with the awesome Vortice rda

 

[druckle]

Guys as always great work, keep it up. Personally, I don't see myself moving away from Ti, even if it is still a bit springy when annealed. Maybe I could see myself trying SS out of curiosity. For now I'll keep enjoying my juice on my DNA200 with the awesome Vortice RDA

NiFe seems to have some real benefits theoretically but alloys are always problematic because of the tight composition controls sometimes needed for reproducibility in critical properties. I think this could be a really important investigation but it also could turn out to be a huge PITA if suppliers don't provide the degree of control we would need for a good material. This is day one of a big undertaking so it's not likely we'll get lucky right off the bat.

If we had our own melting and processing equipment it would make things simple but relying on suppliers can be both time consuming and error prone unless they are primary metals producers who understand the materials they are supplying in detail.

Titanium has the advantage of being a simple pure material so when we buy Grade 1 Ti we know what we're getting. It will probably be more complex with alloys till the kinks can be ironed out.


Duane

 

[TheBloke]

I'm not going to email Crazy now - it's nearly 6pm so they won't see it today.

Instead I'll properly test the TCR at vaping temperature range, which will give us the ultimate assessment of how it compares for vaping. Then I'll email them before the morning.

NiFe seems to have some real benefits theoretically but alloys are always problematic because of the tight composition controls sometimes needed for reproducibility in critical properties. I think this could be a really important investigation but it also could turn out to be a huge PITA if suppliers don't provide the degree of control we would need for a good material. This is day one of a big undertaking so it's not likely we'll get lucky right off the bat.

If we had our own melting and processing equipment it would make things simple but relying on suppliers can be both time consuming and error prone unless they are primary metals producers who understand the materials they are supplying in detail.

Titanium has the advantage of being a simple pure material so when we buy Grade 1 Ti we know what we're getting. It will probably be more complex with alloys till the kinks can be ironed out.

Yeah. What we will be able to do is test a given wire and then approve that given wire for other users. And where it's from a reputable supplier with a specified datasheet, that may well be done for us.

For example, Stealth Vape's NiFe70 is going to have a guaranteed TCR and known resistance. We shouldn't even need to test that to know the figures will be right, and will be the same for all purchasers in future.

ZiVipf is hopefully going to supply Kanthal NiFethal 52, where again we have an approved datasheet which contains specific, presumably guaranteed TCR numbers.

So I think maybe the problems only come when we get an unknown wire, saying it's X% Nickel, Y% Iron or whatever. Like today with Crazy!

 

[druckle]

OK I'll leave out the magnetism

I know trace elements and minor differences in the balance of the metals can vary all these things. I just hadn't realised it could vary quite that much.

LOL Hey Tom this stuff isn't simple. That's why they pay those materials sciences guys the big bucks.

Duane

 

[TheBloke]

I never doubted that

But the point is we still don't have any idea if this wire actually is Ni70Fe30. Maybe it really can vary that much - four times higher resistance than Kanthal NiFethal 70, three times higher than Resistherm, fives times lower TCR etc.. or maybe this is Crazy Wire being Crazy Wire and it's actually Stainless Steel or something.

 

[druckle]

I never doubted that

But the point is we still don't have any idea if this wire actually is Ni70Fe30. Maybe it really can vary that much - four times higher resistance than Kanthal NiFethal 70, three times higher than Resistherm, fives times lower TCR etc.. or maybe this is Crazy Wire being Crazy Wire and it's actually Stainless Steel or something.

Well from experience, +/- 0.02 Yttrium can make the difference whether a turbine blade fails in 100 hours or 20,000 hours. For some things the devil really is in the details.

Duane

 

[druckle]

Tom

I'm fuzzy on this so if you can refresh me I'd appreciate it.

What's the easiest way to measure the TCR of an unknown wire? Your recommendation would be appreciated.
I have a first generation SXK flask and would like to give it a shot on a mystery wire. I looked back through our 70 pages to refresh my memory but haven't come up with the right formula.

Duane

 

[TheBloke]

I think a key thing here is that Crazy specifically said it has 70% Nickel and 30% Iron. They didn't just call it NiFe70 or NiFe30, they said Ni70Fe30.

I've just looked at half a dozen alloys claiming 70%/30%, and they're all similar.

Generic Alloy 120 is 70%/30% and TCR is usually listed as overall TCR around 0.0045, with resistivity around 0.20. A number of places list such an alloy, always with the same/very similar numbers.

Resistherm NiFe30 has 30% Iron (hence being listed as NiFe30, rather than mentioning a '70'), and is about 68.5% Nickel with 1.5% of three other elements. That's the lowest TCR of the ones we've looked at, at 0.0032, with resistivity of 0.33.

Kanthal NiFe70 has one of the higher TCRs, 0.005ish, and is actually 72% Nickel. Ditto Stealth's unknown-make NiFe70, which has an unknown and secret Nickel content somewhere around 70, and is at TCR 0.0049. Both of those have resistivity of 0.20.

So yes I can certainly see that differences in alloys can and do change things. But I can't see how a wire specifically listed as Ni70Fe30, as this is labelled, can have such a different TCR, 0.001-ish, and resistivity - about 0.80.

So at the very least, labelling it "Ni70Fe30" seems like it has to be wrong. If it was truly 70/30, then by definition it would have no other components and it would have to be the same as the Alloy 120s in both resistance and TCR.

Then there's the fact that the resistance, feel, appearance, and result after heating very closely matches stainless steel. Then there's who it's from, a company whom I have recently learnt have quite a reputation for playing fast and loose.

I don't know. I will ask them for a datasheet, but don't expect much. In any case it's all probably moot - if it is a NiFe wire, it's the worst one for vaping by a long shot. It may be fractionally more accurate/sensitive than SS, but only a little, and with the downside of having no idea what's actually in it (given it seems almost certain it can't actually be only 70% Nickel, 30% Iron.) Besides which, much better ones are just around the corner from Stealth and ZiVipf and potentially the mystery US supplier.

 

[TheBloke]

Tom

I'm fuzzy on this so if you can refresh me I'd appreciate it.

What's the easiest way to measure the TCR of an unknown wire? Your recommendation would be appreciated.
I have a first generation SXK flask and would like to give it a shot on a mystery wire. I looked back through our 70 pages to refresh my memory but haven't come up with the right formula.

Duane

I don't know of a method that doesn't require knowing, or at least estimating, the temperature of the wire.

The way I estimated temperature for the quick test with this NiFe was to heat it to glowing. I then estimated, based on what we discussed the other day, that glowing had to be at least 600°C and because it was orange, was more like 700-800°C.

I then watched the live ohms display on the mod while VW firing, making note of the starting ohms and maximum ohms under heating.

I put the figures into this TCR calculator, filling in all fields except TCR which it then calculated for me.

It's crude, because I can't be sure of the max temp. But assuming bright glow = 700°C at least gives you a best case TCR (because it probably heated more than that.)

The proper testing I will do with my thermocouple.

The DNA 200' could well make it easier - easily showing/recording the maximum resistance, which can be compared against the separate thermocouple log showing maximum temperature. Maximum temp and maximum resistance are surely from the same moment, so put those two figures into the TCR calc to get the accurate TCR at that temperature.

Of course that gives the TCR @ 700°C or whatever it heats to unrestricted. We really care about it only up to 250-300°C.

So once one has a rough TCR, switch to TC mode using the rough TCR figure, thermocouple measure the wire with that rough TCR, check if the wire temperature matches the configured temperature on the mod. Adjust the TCR until it does.

The latter is what I did with eg Titanium, where I already knew a rough TCR which I was only validating - 0.0035 in that case, which turned out to be exactly right.

But you might be limited in what you can do without actual temp measuring. But the "glowing hot" test should at least give you a rough idea, whether it's 0.001 or 0.005 or whatever.

EDIT: sorry I'm sick and not thinking straight. Obviously with the thermocouple, one can just record the resistance when it reaches 250°C. As long as one can see what the resistance was at the moment it hit 250°C, one already has the TCR to 250°C

So either timestamped logs from both escribe and thermocouple, or failing that escribe window next to thermocouple log window, and screen capturing software, so one can re-watch the video and see what the resistance was when it hit the target temp.

One could also heat it gently, say 10W, with airflow, to minimise the heating so it doesn't overshoot

Shouldn't be a need to re-tune the TCR after that.

 

[TheBloke]

Here's an even cruder way. All the following assume the base coil was @ 20°C / 68°F

• An Ni200 coil @ approximately 0.006 TCR will increase its resistance 2.35 times by 250°C (0.2 -> 0.47), and by 3.88 times by 500°C
• A wire with 0.003 TCR will increase its resistance 1.69 times by 250°C (0.2 -> 0.338), and 2.44 times by 500°C (0.2 -> 0.488). It will have tripled by 686°C
• A wire with 0.001 TCR will increase its resistance 1.23 times by 250°C (0.2 -> 0.246)
• A wire which has doubled its resistance by 600°C has TCR 0.0017

So VW heat it up gently. If the resistance hasn't at least doubled by the time it's glowing, it's definitely a low TCR. Ideally if you're heating to glowing you want at least a 3x resistance rise, then it might be at least 0.003.

Depends on the base resistance of course. If you find a wire where an 8 x 3mm x 26G coil is 1.0Ω and it has 0.001 TCR, it might actually be quite sensitive/accurate. A lot more than SS anyway.

 

[druckle]

I don't know of a method that doesn't require knowing, or at least estimating, the temperature of the wire.

The way I estimated temperature for the quick test with this NiFe was to heat it to glowing. I then estimated, based on what we discussed the other day, that glowing had to be at least 600°C and because it was orange, was more like 700-800°C.

I then watched the live ohms display on the mod while VW firing, making note of the starting ohms and maximum ohms under heating.

I put the figures into this TCR calculator, filling in all fields except TCR which it then calculated for me.

It's crude, because I can't be sure of the max temp. But assuming bright glow = 700°C at least gives you a best case TCR (because it probably heated more than that.)

The proper testing I will do with my thermocouple, and potentially a camera to record the temperature alongside the resistance - unless I can just eyeball that well enough.

The DNA 200' could well make it easier - easily showing/recording the maximum resistance, which can be compared against the separate thermocouple log showing maximum temperature. Maximum temp and maximum resistance are surely from the same moment, so put those two figures into the TCR calc to get the accurate TCR at that temperature.

Of course that gives the TCR @ 700°C or whatever it heats to unrestricted. We really care about it only up to 250-300°C.

So once one has a rough TCR, switch to TC mode using the rough TCR figure, thermocouple measure the wire with that rough TCR, check if the wire temperature matches the configured temperature on the mod. Adjust the TCR until it does.

The latter is what I did with eg Titanium, where I already knew a rough TCR which I was only validating - 0.0035 in that case, which turned out to be exactly right.

But you might be limited in what you can do without actual temp measuring. But the "glowing hot" test should at least give you a rough idea, whether it's 0.001 or 0.005 or whatever.

EDIT: sorry I'm sick and not thinking straight. Obviously with the thermocouple, one can just record the resistance when it reaches 250°C. As long as one can see what the resistance was at the moment it hit 250°C, one already has the TCR to 250°C

So either timestamped logs from both escribe and thermocouple, or failing that escribe window next to thermocouple log window, and screen capturing software, so one can re-watch the video and see what the resistance was when it hit the target temp.

One could also heat it gently, say 10W, with airflow, to minimise the heating so it doesn't overshoot

Shouldn't be a need to re-tune the TCR after that.

OK thanks Tom. I don't have decent temp measuring capability but I'll see what I can accomplish with the DNA 200. Do you think the SXK and a simple water test get me in the general range if I search for an indicated water boiling point with various "purity" settings?

Hope you feel better soon. I used to have lots of allergy issues in Connecticut. When I moved to Arizona they went away but for most folks the desert species pollens etc. are worse than the East Coast. I got lucky.

Duane

 

[TheBloke]

Yeah I always forget about that, if it hovers at 100°C/212°F then it's at the right TCR - right?

I'd still do the glowing test first to get you a ballpark, that seems quicker to me because in one test - just watching the screen on the SXK - you can immediately see if it's 0.001 or 0.005 and then tune in the right range for the next tests. It'd only be inaccurate if it the TCR is massively non-linear. But I think the only wire vaguely matching that description is Ni200/Kanthal DH, that we've seen anyway. In any case at a low wattage you can get an idea of the progression of resistance and get the ballpark.

And thanks!

 

[TheBloke]

I just did a "slow burn" test on the dreaded Crazy 'NiFe', heating only at 5W. I thought maybe this could be a non-linear one, so I wanted to see if all the resistance rise came at the start of the heating. Then I thought I may as well just do it properly and got the Thermo out. I didn't do my full normal test harness setup, just quickly wedged the probe between two wraps. If anything that will over-estimate the TCR (under estimate heating.)

At about 300°C the resistance rose from 0.47 to 0.60, so a TCR of around 0.00108. So, yeah. No need for proper tests I think.

(That TCR is almost identical to SS304 @ 300°C, within the margin of error of my quick test anyway. Perhaps just a coincidence.. )