|
Homebrew Equipment A place to show, design and discuss the weird and wonderful electronic creations from the hands of individual members. |
|
Thread Tools |
20th Aug 2013, 10:11 am | #21 |
Heptode
Join Date: Feb 2003
Location: Garnant, near Ammanford, South Wales, UK.
Posts: 657
|
Re: Variac affecting moving-iron meter
How about fitting a digital meter
__________________
BVWS member. |
20th Aug 2013, 11:16 am | #22 |
Nonode
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
|
Re: Variac affecting moving-iron meter
I like your lateral thinking! That would be too easy, though. I simply fancied the 1950s-style look of these panel meters, and I've done the hard bit (the metalwork) now, so I'm determined to get them to work.
Chris
__________________
What's going on in the workshop? http://martin-jones.com/ |
20th Aug 2013, 8:37 pm | #23 |
Nonode
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
|
Re: Variac affecting moving-iron meter
I've now tried making a screen using 1.6mm steel sheet, bent round into a cylinder to fit round the meter. It improved the accuracy still further, though there's still a detectable influence from the variac. I haven't tried making a screen which covers the back of the meter as well, though. Maybe I will if I'm feeling very enthusiastic about metalwork!
Interestingly it made no difference whether I welded the ends of the steel strip together or not. Chris
__________________
What's going on in the workshop? http://martin-jones.com/ |
21st Aug 2013, 10:14 am | #24 |
Nonode
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
|
Re: Variac affecting moving-iron meter
I tried adding a screen (just tinplate cut out with tinsnips) to the back of the meter as well as the sides, and it made a useful difference, so I decided to keep it. I made another screen for the voltmeter because that was also being affected: changes in load on the variac made the pointer swing all over the place. I attach a picture of the end result. Sorry about the awful welding. I've run out of gas for the MIG, so I had to weld without any shielding gas. The tinplate back is just soldered on to the cylindrical screen. There are two holes drilled and tapped in the sides with screws in which can be tightened to gently grip the meter casing and hold the screen in place.
Now both meters have a tendency to over-read by about 5%, but they are dramatically better than they were without the screens. Lesson learned! Keep your moving-iron meters away from Variacs, or make sure they're magnetically screened. Thank you everyone for the suggestions and advice. Chris
__________________
What's going on in the workshop? http://martin-jones.com/ |
21st Aug 2013, 1:52 pm | #25 |
Rest in Peace
Join Date: Apr 2011
Location: Matlock, Derbyshire, UK.
Posts: 1,378
|
Re: Variac affecting moving-iron meter
I don't know whether it will make much difference but you could twist the meter cables together to reduce magnetic pick-up.
|
21st Aug 2013, 11:01 pm | #26 |
Nonode
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
|
Re: Variac affecting moving-iron meter
I tried moving the cables around - it makes a difference, but only if they're actually wrapped around the case of the meter.
__________________
What's going on in the workshop? http://martin-jones.com/ |
4th Nov 2013, 9:04 am | #27 | |
Retired Dormant Member
Join Date: Feb 2008
Location: Auckland, New Zealand
Posts: 631
|
Re: Variac affecting moving-iron meter
Quote:
Shielding can be difficult to implement because thin (easily workable) materials saturate at low field levels and thus their shielding effect is minimal. Multi-layer steel-aluminium-steel shielding material offers an improvement and works very well, with the eddy currents in the aluminium core adding to the field cancellation effects. I work on field mitigation projects for Universities and Hospitals etc and by far the best strategy is to move the source away from the affected instrument then use shielding to mop up residuals, followed by active field cancellation if necessary. The latter was the only technology ever to make my principal field meter read zero nanotesla in all three axis simultaneously. In this instance it is inevitable that the meter will be affected by the core fields, and of course the interference will worsen as the variac load increases. Cheers Billy |
|
4th Nov 2013, 11:19 am | #28 |
Rest in Peace
Join Date: Apr 2011
Location: Matlock, Derbyshire, UK.
Posts: 1,378
|
Re: Variac affecting moving-iron meter
The core field will be almost constant regardless of the load. There will of course be extra field generated by load current external to the Variac especialy if the loop is large.
|
26th Dec 2013, 7:05 am | #29 | |
Retired Dormant Member
Join Date: Feb 2008
Location: Auckland, New Zealand
Posts: 631
|
Re: Variac affecting moving-iron meter
Quote:
Cheers Billy |
|
26th Dec 2013, 11:56 am | #30 |
Dekatron
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 4,923
|
Re: Variac affecting moving-iron meter
As the years go by , I am increasingly inclined to regard "weight", as in "lack of weight" as a very important technical parameter for electronic equipment. My 8amp Variac weighs quite enough as supplied, so when I wanted to add meters to it, the negligible weight of an inexpensive digital meter, reading both V & A, was irresistible. It had not occurred to me when making that decision, that I would also avoid the problem which magnetic fields have on other types of meters .
|
26th Dec 2013, 2:42 pm | #31 |
Dekatron
Join Date: Apr 2012
Location: Worksop, Nottinghamshire, UK.
Posts: 5,549
|
Re: Variac affecting moving-iron meter
I have found those square moving iron meters that used to be sold by RS to be fine even hard up against the Variac coils.
|
26th Dec 2013, 7:26 pm | #32 |
Nonode
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
|
Re: Variac affecting moving-iron meter
I think the square meters have steel cases, which would make them more tolerant of the field from the variac. My round ones just have bakelite cases. They were very cheap, though...maybe that's one reason why!
Chris
__________________
What's going on in the workshop? http://martin-jones.com/ |
28th Dec 2013, 12:13 am | #33 |
Dekatron
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,526
|
Re: Variac affecting moving-iron meter
I still don't see that the magnetic field in the variac core increases with load current. Apart from anything else, if a TX is well designed and uses the core close to saturation, any field increase would saturate the core, which doesn't happen!
A variac is no different from an ordinary transformer with one end of the secondary connected to one end of the primary. Any change in core field caused by secondary loadcurrent is compensated for by a corresponding change in the primary current, leaving the core field at the same level- at least to a first order, leaving out any non ideality of the transformer. Magnetic fields round any connecting wires or loops will of course vary with load.
__________________
....__________ ....|____||__|__\_____ .=.| _---\__|__|_---_|. .........O..Chris....O |
3rd Jan 2014, 9:08 am | #34 | |
Retired Dormant Member
Join Date: Feb 2008
Location: Auckland, New Zealand
Posts: 631
|
Re: Variac affecting moving-iron meter
Quote:
Field meter was a professional triaxial & true RMS FD1 by Combiniva of Sweden. Cheers Billy |
|
3rd Jan 2014, 9:23 am | #35 |
Moderator
Join Date: Jan 2003
Location: Ipswich, Suffolk, IP4, UK.
Posts: 21,192
|
Re: Variac affecting moving-iron meter
What are the units on the X axis?
__________________
Graham. Forum Moderator Reach for your meter before you reach for your soldering iron. |
3rd Jan 2014, 10:14 am | #36 |
Nonode
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
|
Re: Variac affecting moving-iron meter
That's very interesting indeed! Graham, I think the units on the X axis are the cells in the table at the top, so they're roughly 'voltage output' but non-linear. It's nice to see some hard data that shows I'm not going mad - the field strength really does vary with variac output, and in a non-obvious way, too. I also suspect it varies in different ways at different points around the variac, since the current in the two 'halves' of the winding varies both in magnitude and location as the wiper moves.
__________________
What's going on in the workshop? http://martin-jones.com/ |
3rd Jan 2014, 3:38 pm | #37 |
Rest in Peace
Join Date: Apr 2008
Location: London, UK.
Posts: 2,508
|
Re: Variac affecting moving-iron meter
To concur with and expand upon the previous post (which I had missed somehow!) it's interesting but not surprising - the S-shape is probably explained by the simple fact of the input being connected to a tap in the winding rather than the finish. But it's worth exploring why the curve between the minima is of that form...
You don't mention along which axis the measurement is taken or at what circumferential position relative to the ends of the Variac winding. However those are most important parameters because leakage is anything but geometrically uniform.The variac 'setting' varies not only the electrical transformation ratio but also the physical shape of the active parts of the magnetic circuit, therefore by inspection some components of the leakage flux will change not only in magnitude but also in spatial vector according to the variac setting. We also apparently have to contend with varying load current, if the power is a constant 200W. Let's consider the effects of these parameters. To start with, I would be tempted to discount the effect of the coiled power cord as it's a bifilar winding, and air cored. If the load power is constant at 200W, we are varying not only the voltage but also the secondary current as an inverse function of voltage. I am not sure how you have dealt with the discontinuity approaching zero volts when the load current would exceed the Variac rating? To describe the spatial effect of the setting, consider first a hypothetical variable voltage device consisting of a layer-wound autotransformer of very many turns on a single-bobbin EI core, with each turn brought out to a tapping selector of arbitrarily many ways. Electrically it works like a Variac, but magnetically it will behave like any typical EI transformer, with leakage MMF along any particular external vector varying mainly as a function of the flux density in the unwound part of the core. Hence with the relatively constant magnetising current, one might expect only a modest variation in leakage due to the loss factors, and in approximately scalar fashion. Compare a Variac, in which the physical disposition of the turns passing load current varies with brush position. At minimum setting it is an unloaded toroid with a 'bald patch' between start and finish, leading to fringe effects around the ends of the winding. At maximum it is the same. At 50% the winding on half the core is carrying input current only, the other the vector sum current. Discounting the 'bald patch', the layout now approximates to a toroid with a winding covering only half the circumference, carrying the sum current. One can imagine this progressing around the core as the setting is altered, carrying with it certain components of the leakage while others remain pegged relative to the 'bald patch'. Now, looking at the curve above, the maximum voltage exceeds normal mains voltage, therefore I assume the supply is connected to a tap in the winding. The minima appear to coincide with settings of zero, and where the wiper is located at the likely position of the input tap. These are the cases mentioned above as 'minimum' and 'maximum'. For settings where the brush is not at either end of the current-carrying section of the winding, the leakage flux at the measurement point increases. But if you were to measure at many points around the core, I expect you would see a wide variety of different curves. With delta changes in setting they will perhaps all rotate through fractions of the setting angle, but with large changes you might obtain curves of different orders entirely, their form differing also according to the V/I curve of the load. They would be interesting to study! Last edited by Lucien Nunes; 3rd Jan 2014 at 3:56 pm. Reason: Addition |
22nd Jan 2014, 6:57 am | #38 | |
Retired Dormant Member
Join Date: Feb 2008
Location: Auckland, New Zealand
Posts: 631
|
Re: Variac affecting moving-iron meter
Quote:
The insurance company changed its mind, accepted my client's claim and paid my account into the bargain. Cheers Billy |
|