Shopping for a variac

[drussell]

Anyway, the reason why is because a fuse generally starts thinking about melting at around 50% of its rating.

Huh?

That's not how fuse ratings work.

A typical Littelfuse 3AG Fast-Blow fuse is rated to hold 100% of rated current for a minimum of 4 hours continuous.

On the other hand, at 135% of rating it must open in less than 1h maximum.  At 200% it must open in seconds.

The input fuse still has virtually nothing to do with the actual output current rating and allowed load characteristics.

Edit:

But anyone with the least amount of electrical knowledge should know that your limitation is around 2.5 amps

[Cubdriver]

If they meet their specs, then isn't that an adequate design? Why do you presume these are inadequate designs? I'm just looking for facts, not vague comparisons to "big boys", whatever that is.

LOL...  That's just it!

They don't even HAVE specs!

A proper Variac like the 7.5A Powestat models above will be able to do it's full rated output current of 7.5A, even on the very bottom end of the winding.

The Chinese ones don't even tell you how much current they're supposed to be able to deliver.

They tell you exactly. There's a big 5AMP fuse label on the front. Which means you're limited to 2-3 amps.

And do you know how old those Powerstats are? I'm guessing 1940's. That company has been around since the 1930's. It's up to you, but buying a device that's been kicked around for maybe 70 years doesn't seem like a good choice to me.

So, the 5A fuse means that they're good for 2-3 amps?  Exactly?     2-3A doesn't seem very 'exact' to me - seems a rather large margin, actually.  You buy your cheap but shiny and new far east variac, and I'll stick with old but dull used ones in decent condition with a proven track record and conservative specs that they will meet day all day every day.

-Pat

Come on, this isn't rocket science. You can go to the Littlefuse website, find a 5 amp fuse, and look at the time/current curve. It tells you exactly the time it takes to melt at any current. But anyone with the least amount of electrical knowledge should know that your limitation is around 2.5 amps, and if it's that important you can test it and buy another fuse.

So to determine the actual output capabilities of my el-cheapo variac, I need to go to the FUSE MANUFACTURER'S website, look up the fuse, and then do calculations?  Seriously?   

<Edit to add - and is the far east built-to-be-as-cheap-as-possible variac going to HAVE a Littelfuse or Bussman fuse that is accurately characterized in it?  Or will it be a Wun Hung Lo brand fuse of who knows what actual rating?>

[bdunham7]

They tell you exactly. There's a big 5AMP fuse label on the front. Which means you're limited to 2-3 amps.

And do you know how old those Powerstats are? I'm guessing 1940's. That company has been around since the 1930's. It's up to you, but buying a device that's been kicked around for maybe 70 years doesn't seem like a good choice to me.

Holy carp dude!  Do you have a warehouse full of these things that you're selling?  Your wild guesses are no better than anyone elses here.

First, I'll admit that my estimation of the ampacity of the 24awg winding was off.  I based it on a transformer winding rule-of-thumb of 2.3A/mm^2 and then doubling it because it's a variac.  I looked at two variacs of mine, both Staco, one with 24AWG rated at 3A and one with 26AWG rated at 2.25A.  So reputable manufacturers are using about 6.0A/mm^2.

However, I don't see how you justify your assertion that they tell you exactly that the maximum current is 2-3 amps (as if "2-3 amps" is exact in the first place) especially since it says 500VA on the front as well.  I would argue that that labeling would permit you to connect a 5A load at 100V and let it roast. 

Lastly, how much obvious QC will you ignore in your support of this junk.  Poor solder, wiring internally wrong, not grounded, etc, and you think you can just fix all this (on a brand new unit) and just assume that all of the other things that you can't test or didn't notice will be just fine?  That the enamel insulation is top quality, the iron core is sufficient, etc etc?  When a product or company has severe, obvious issues the first thing I think is "what else did they screw up" not "gee, I can patch that up and I'll have a bargain".  If I'm going to patch something up, I'd prefer a high quality product that has withstood the test of time.

[Cubdriver]

They tell you exactly. There's a big 5AMP fuse label on the front. Which means you're limited to 2-3 amps.

And do you know how old those Powerstats are? I'm guessing 1940's. That company has been around since the 1930's. It's up to you, but buying a device that's been kicked around for maybe 70 years doesn't seem like a good choice to me.

Holy carp dude!  Do you have a warehouse full of these things that you're selling?  Your wild guesses are no better than anyone elses here.

First, I'll admit that my estimation of the ampacity of the 24awg winding was off.  I based it on a transformer winding rule-of-thumb of 2.3A/mm^2 and then doubling it because it's a variac.  I looked at two variacs of mine, both Staco, one with 24AWG rated at 3A and one with 26AWG rated at 2.25A.  So reputable manufacturers are using about 6.0A/mm^2.

However, I don't see how you justify your assertion that they tell you exactly that the maximum current is 2-3 amps (as if "2-3 amps" is exact in the first place) especially since it says 500VA on the front as well.  I would argue that that labeling would permit you to connect a 5A load at 100V and let it roast. 

Lastly, how much obvious QC will you ignore in your support of this junk.  Poor solder, wiring internally wrong, not grounded, etc, and you think you can just fix all this (on a brand new unit) and just assume that all of the other things that you can't test or didn't notice will be just fine?  That the enamel insulation is top quality, the iron core is sufficient, etc etc?  When a product or company has severe, obvious issues the first thing I think is "what else did they screw up" not "gee, I can patch that up and I'll have a bargain".  If I'm going to patch something up, I'd prefer a high quality product that has withstood the test of time.

He seems to like taking contrarian positions then digging in hard.

https://www.eevblog.com/forum/testgear/good-power-supply/msg3019052/#msg3019052

https://www.eevblog.com/forum/testgear/atx-power-supplies/msg3025932/#msg3025932

-Pat

[engrguy42]

bdunham, I'm not supporting anything. I don't give a flying F@@@ what anyone buys. But for some reason you guys can't comprehend that I'm looking for facts. I don't have personal grudges, I'm not a fanboy for anything, I'm just doing what engineers do: sort thru the personal bias and unsupported BS generalizations and find the facts.

Apparently that type of approach pisses people off here because it challenges you to put up or shut up. And since you can't provide data you resort to personal attacks.

Fine. When somebody can provide facts they're worth listening to. Otherwise it's just the typical muscle flexing to act smart. 

[tkamiya]

Guys?

This is why I volunteered to test this thing.  Too much theorizing over things we don't understand 100%.  Everyone knows bits and pieces. There's theory and there's what works in practice.  Has everyone seen very large slidac like 3 phase 200 amp job?  I have.  It was motorized.  Then had a fun discussion with an old instructor. 

Today, I'm feeling very poorly.  I don't know if I can get it done today but I am curious.  I wanna see this thing spark and burn up.  I have 3 fire extinguishers loaded on a cart.  Face shield is ready. 

Someone asked me about the lens.  I have 24-70 2.8, 70-210 2.8, 150mm 2.8 macro, 8mm fish eye, etc, etc, etc.  One of the deficiencies of D800 was focus while shooting.  That's what D810 solves.  I think.  I never used it for video.  This is the first time.

[tkamiya]

By the way, I did study fuse and breakers in depth when I was in trade school.  5A fuse will pass 5A.  But good practice is if you need to pass 5A, better use bigger fuse.  Breakers work little differently.  You need to pass 5A, you use 5A.  But usual usage needs to be 70 to 80% of that.  Otherwise you have to up-size everything.  I can dig up an old text book with rating curve and all that but I'm not going to translate it.  So start studying Japanese 

This thread is starting to resemble a lug nut thread on automotive forum.

[drussell]

He seems to like taking contrarian positions then digging in hard.

Indeed, all the while talking about only wanting the facts. 

It seems that he is one to like to latch onto whatever "facts" support his particular preconceived position, like if a Variac says it should have a 5A fuse in it, that must mean it is rated for an output of "2-3A" (is that resistive only?  who knows), while essentially dismissing any and all evidence to the contrary.

I'm now trying out the user-ignore feature for some serenity. 

[engrguy42]

By the way, I did study fuse and breakers in depth when I was in trade school.  5A fuse will pass 5A.  But good practice is if you need to pass 5A, better use bigger fuse.  Breakers work little differently.  You need to pass 5A, you use 5A.  But usual usage needs to be 70 to 80% of that.  Otherwise you have to up-size everything.  I can dig up an old text book with rating curve and all that but I'm not going to translate it.  So start studying Japanese 

This thread is starting to resemble a lug nut thread on automotive forum.

Yes, that was my mistake. I was thinking larger power fuses. I can give examples but the whole discussion is irrelevant.

[Cubdriver]

bdunham, I'm not supporting anything. I don't give a flying F@@@ what anyone buys. But for some reason you guys can't comprehend that I'm looking for facts. I don't have personal grudges, I'm not a fanboy for anything, I'm just doing what engineers do: sort thru the personal bias and unsupported BS generalizations and find the facts.

Apparently that type of approach pisses people off here because it challenges you to put up or shut up. And since you can't provide data you resort to personal attacks.

Fine. When somebody can provide facts they're worth listening to. Otherwise it's just the typical muscle flexing to act smart.

So where's the data for the shiny red variac?  Someone posted a catalog from Superior Electric showing the specs of the Powerstats.  Current ratings, temperature rise, construction points, connection diagrams  That's data that, as an engineer, I can look at and be confident in.  On the other hand, the shiny red variac has???  The "data' we're providing is that the established brands actually PUBLISH data, and the cheapies do not.  How do I compare data to no data?  If I'm buying something and manufacturer A provides me a data sheet showing characteristics, and option B is something on amazon that is sold by multiple outlets with a possible name change or different colored case and some sort of vague specs that it *might* meet, well, to me the choice is pretty obvious.

-Pat

[bdunham7]

bdunham, I'm not supporting anything. I don't give a flying F@@@ what anyone buys. But for some reason you guys can't comprehend that I'm looking for facts. I don't have personal grudges, I'm not a fanboy for anything, I'm just doing what engineers do: sort thru the personal bias and unsupported BS generalizations and find the facts.

Apparently that type of approach pisses people off here because it challenges you to put up or shut up. And since you can't provide data you resort to personal attacks.

Fine. When somebody can provide facts they're worth listening to. Otherwise it's just the typical muscle flexing to act smart.

What qualifies as a 'fact' for you?  Are the conclusions you have asserted supported by such facts?

[drussell]

Someone asked me about the lens.  I have 24-70 2.8, 70-210 2.8, 150mm 2.8 macro, 8mm fish eye, etc, etc, etc.  One of the deficiencies of D800 was focus while shooting.  That's what D810 solves.  I think.  I never used it for video.  This is the first time.

With a D or G series lens installed (and possibly other newer ones, I don't know, and their are proper dedicated cine lenses too, of course) you should theoretically have options in the Auto-Focus setup while you're in Movie mode to have the AF configured between face-priority, wide-area AF, normal-area AF, subject-tracking, full-manual, etc.

What does it list for possible options in the Auto-Focus setup, in Movie mode, with the lens you're using installed?

Barring that, are you possibly running very old firmware, is there perhaps a firmware update?

[engrguy42]

Guys?

This is why I volunteered to test this thing.  Too much theorizing over things we don't understand 100%.  Everyone knows bits and pieces. There's theory and there's what works in practice.  Has everyone seen very large slidac like 3 phase 200 amp job?  I have.  It was motorized.  Then had a fun discussion with an old instructor. 

Today, I'm feeling very poorly.  I don't know if I can get it done today but I am curious.  I wanna see this thing spark and burn up.  I have 3 fire extinguishers loaded on a cart.  Face shield is ready. 

Someone asked me about the lens.  I have 24-70 2.8, 70-210 2.8, 150mm 2.8 macro, 8mm fish eye, etc, etc, etc.  One of the deficiencies of D800 was focus while shooting.  That's what D810 solves.  I think.  I never used it for video.  This is the first time.

Sorry to hear you're not feeling well. Don't bother with the testing. Sadly, I think this is yet another situation where people believe what they want to believe, and facts are irrelevant. I'm thinking I might also buy one and do my own testing to see what the actual design limitations are, if any. More data is better. 

[Cubdriver]

Buy one, open it up carefully, take a look inside, give it a test, and go from there

If the usual suspects don't give you grief to fix, call it a keeper for now

Otherwise find the receipt if bought local, return it,
and let the next guy cop it, and learn the hard way that going too cheap on such gear is a NO GO 

I 100% agree with the idea of "don't buy junk". However, these devices are so simple that, aside from serious design flaws, getting them working well seems to be pretty simple. You can always point to a "yeah, but...". But this is different from a complex scope, for example, that you can't repair with a quick soldering iron. And the benefit of saving $200+ might outweigh the cost of some minor repairs.

The big question (for me at least) is whether there are any serious design flaws that apply to many/most units.

Reading back a little bit, this one struck me.  Let's go with your assertion that "these devices are so simple".  If they're 'so simple', then why can't the MANUFACTURER make them right?  Why should someone NEED to open up their brand new shiny variac to make sure that it's grounded properly, or otherwise wired correctly?  They're such simple devices, you'd think the place that makes them could do it.  The big question to ME is that if the place that builds them can't consistently PUT THEM TOGETHER CORRECTLY, how can I be confident that they are designed properly?  Looking at the photos put up earlier, it appears to me that the Powerstat I posted has a brush with about 4x the contact area as the conical one shown on the red variac.  The Powerstat is rated at 3A and the instructions are to fuse it at that level.  The red one is fused at 5A.  Is the red one adequately designed and Superior Electric so super conservative that they used 4x brush size for 3/5 of the current, or is the red one running right at the edge?  As an engineer, look at the two and tell me which one YOU think will be more reliable in the long term.

-Pat

[tkamiya]

Starting with slightly modified process.  I'll be back shortly.  Having issue with video.

[engrguy42]

Damn...4 thermocouples. Excellent !!!

Just be careful. I know you have a fire extinguisher, but be careful. Think 3 times before doing anything 

[engrguy42]

BTW, I'm curious what folks are expecting for temperatures. What constitutes a "fail"?

[drussell]

BTW, I'm curious what folks are expecting for temperatures. What constitutes a "fail"?

We're looking for data, not a pass/fail go/no-go test.

We would like to know things like core temperature rise after running at a given load long enough to stabilize, etc. so that it can be compared to the known data from other units' known, published specifications.  eg. those 1958 PowerStat specs state there will be no more than 50°C core temperature rise when operated at maximum ratings (which is maximum current down low or the total maximum allowed VA, depending on wiper position), ambient temperature -20 to +40°C, with a published de-rating curve for operation above 40°C ambient.

etc.

[Shock]

My moneys on the Fluke surviving.

[engrguy42]

BTW, I'm curious what folks are expecting for temperatures. What constitutes a "fail"?

We're looking for data, not a pass/fail go/no-go test.

We would like to know things like core temperature rise after running at a given load long enough to stabilize, etc. so that it can be compared to the known data from other units' known, published specifications.  eg. those 1958 PowerStat specs state there will be no more than 50°C core temperature rise when operated at maximum ratings (which is maximum current down low or the total maximum allowed VA, depending on wiper position), ambient temperature -20 to +40°C, with a published de-rating curve for operation above 40°C ambient.

etc.

So it seems like the goal is to end up saying "See, the old Powerstat is better because it has lower temperatures"? I'm not sure how, or to what extent, a comparison to other equipment is relevant.

To me it seems the question is "Can these cheap Chinese variacs supply around the rated 5Amps/500VA continuously without overheating and causing internal damage" or something like that. 

[tkamiya]

OK, done. 

Summary:
Environment is outside in Florida 90 to 100 degrees F depending on if sun is out or not.
Total run time 10 minutes
50V 2.4amp for 8 minutes or so
120V 4.2amp for rest of the test
Highest 120V 4.2A = 504VA = 504watts
Core temperature stayed around 100 degrees F.
Hottest part is 137 F.  The big nut at bottom which holds the spindle.
Side and top is somewhere between the two.
No spark, nothing melting, no smoke.

At this time, I don't see any reason to continue so ended the test around 15 minutes.
Messing with video right now but do you guys really want to see this?  Absolutely nothing happens.

[engrguy42]

OK, done. 

Summary:
Environment is outside in Florida 90 to 100 degrees F depending on if sun is out or not.
Total run time 10 minutes
50V 2.4amp for 8 minutes or so
120V 4.2amp for rest of the test
Highest 120V 4.2A = 504VA = 504watts
Core temperature stayed around 100 degrees F.
Hottest part is 137 F.  The big nut at bottom which holds the spindle.
Side and top is somewhere between the two.
No spark, nothing melting, no smoke.

At this time, I don't see any reason to continue so ended the test around 15 minutes.
Messing with video right now but do you guys really want to see this?  Absolutely nothing happens.

Good stuff. Thanks. I'm kinda surprised the temps are so low (137F/58C) , but I suppose with the outdoor breeze maybe that helped.

[drussell]

OK, done. 

Summary:
Environment is outside in Florida 90 to 100 degrees F depending on if sun is out or not.
Total run time 10 minutes
50V 2.4amp for 8 minutes or so
120V 4.2amp for rest of the test
Highest 120V 4.2A = 504VA = 504watts
Core temperature stayed around 100 degrees F.
Hottest part is 137 F.  The big nut at bottom which holds the spindle.
Side and top is somewhere between the two.
No spark, nothing melting, no smoke.

Well, at 120v it is just going straight through....  so that's kinda pointless to check

The highest voltage drop should be around 60v and also at the highest boost setting.

The biggest question is can you actually pull 4 or 5 amps out of it at those "tougher" settings, where it will be dissipating the most, or right at the bottom of the range, etc. 

You will need additional load than just the single bulb to test that, though....

Edit:



I would like to know how that tiny brush behaves when you crank it back and forth with 4 or 5 amps through it, too.

[engrguy42]

drussell, are you concerned about some resistive dissipation in the windings causing higher temps?

[tkamiya]

LCD is just plain awful outside.  Can't read a thing!  So I don't think I'll upload this.  Is that ok with you guys?  All you get is for 16 minutes worth of what I just summarized on previous post.

Yes, breeze definitely helped.  But high ambient temp did not.  So this is not a conclusive test.  There is no way I'd do this inside or in a box of some kind. 

As to how to interpret this, I am really not sure.  I am still not comfortable leaving it unattended.  I think any slidack by anyone will see some rise in temp.  Was it excessive?  I don't think so.  I think it was reasonable.  How it was constructed leaves me with a lot of reservations.  So from here, unless someone does more test and more accurate test, it will just be personal feelings.

So end result is:
I'm alive.
My house did not burn down.
Fuse survived.
No sparks.
Moderate temp rise.