Go Back   Home > Forums > General Interest > Everything Else

Everything Else Anything related to audio / video / electronics etc) BUT remember- we have many new forums where your thread may now fit! .... Parts, Equipment & Tools, Construction Tips, Software Tools......

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 30th January 2008, 01:41 AM   #1
poobah is offline poobah  United States
diyAudio Member
 
poobah's Avatar
 
Join Date: Nov 2005
Default Calling jneutron, EVA. SY, precision high power resistor

OK...

I have built an invertor... oops let's call it a 60 Hz class D amplifier. At any rate, it is fairly efficient... in fact my first pass at measuring efficiency yielded a result of about 101%.

Normally... the poo would proudly beat his chest... and mail an invoice. I have made the mistake in this case of informing my client that 101% efficiency is probably not real.

The problems are obvious... one new 6.5 digit HP meter is here. The rest of the meters are at the cal lab right now...

Now... while I appreciate the ideas of calorimeters and the that junk... reality has to apply here as well.

So... one of the ways to eliminate variables is to use a fixed load resistor and measure the output RMS to calc. power out... all good.

I have 100 meters each of 0.05mm & 0.125mm manganin wire on the way. These are about 0.35 and 2.19 ohms per meter irrespectively.

The plan:

I plan to immerse these in a bath of food grade mineral oil... 18 "meg" or better. The bath will consist of a 5 gallon bucket(?), a temp probe, and pond pump or something. I can perhaps concoct some inner bucket/weir to force a nice fluid-flow loop.

The goal of said flow loop is keep the oil/wire temperature homogenous.

So... after I build this thing (yes... with Kelvin connections) I will take IT to the cal lab and have the resistance of various branches measured over temperature... so that I can fit a function and characterize them... blah blah.

THE QUESTION(s):

1. How should I wind this thing to minimize/avoid inductance?

2. Any bright ideas on coil forms? Please no real-high dollar solutions... McMaster Carr ideas are fine... and something better than a wicker wastebsket is probably justified.

OH... BTW, THD + N is running at about 0.23%... so, we don't have a lot of harmmonic trash to resonate in strays... or throw off the RMS-ser in the Voltmeter.

The output of the inverter is 100 Watts 24 VAC 60 Hz. Power is not a huge concern... it should really only be limited by fluid flow, fluid volume, and the melting temp of the resistor vessel (pickle bucket?) Test durations are not long anyway... I have PLENTY of loads for duration testing etc... Trying to measure efficiency... for a published spec... to +/- 0.1% says we need +/- 0.02% on contributing readings... easy to SAY.

Food fights are welcome...

And cool it on the buy-3-more-meters-thing too. That puts 2 meters in an RMS * RMS thing... bad. I need a stable high power resistor for this project ongoing. After hearing about the 101% efficiency... the client is asking for more models and sizes.

$50,000 for a new BURWICKEE & LIMPCOCKE AC analyzer won't do either. The client pays either way... I would rather the 50K went to The Poo Foundation for Drunken Women with Low Self-Esteem.

  Reply With Quote
Old 30th January 2008, 02:05 AM   #2
diyAudio Member
 
Join Date: Mar 2007
Location: Canandaigua, NY USA
They told me I was crazy to wind my own resistors, but as you can see, it solves a lot of problems. So you need something like 5.76 ohms that's stable at 100W. Should be easy with a small pail of oil. There are all sorts of published winding schemes, but I've found most aren't that good for inductance, not to mention a big PITA to wind. For low values, a bit of capacitance doesn't matter. Thus, I use the bight method. Fold the length of wire in half and keep the pair together when winding. I like something flat- it's asking for trouble to use a round form, though it will work. Traditionally, a brass tube was often used. Maybe a sheet of glass with the edges stoned smooth and/or taped with Kapton tape? You can twist the wires together a bit- this will help with the inductance, but you don't want an insulation failure. Unlikely at 24VAC. Now, if the wire isn't insulated, you have a bigger problem. You'll want a sheet of something with small sawn or filed notches to keep the wire orderly, and still run a folded pair. Not sure why you have to send it out, as with calibrated Agilent meters, you can do a super 4-terminal measurement that's better than you need. Note that you should bake your resistor a bit to stabilize it before use. Easy to do- just power it up, not smokin hot, but above the expected operating temperature.
  Reply With Quote
Old 30th January 2008, 03:11 AM   #3
poobah is offline poobah  United States
diyAudio Member
 
poobah's Avatar
 
Join Date: Nov 2005
Thanks Conrad,

When you say twist the wires... Do you mean both halves? After the fold?

Yes 5 Ohms or even much less is the target. If I'm going to go through this much grief, I might as well build for the future req's.

And yes, I failed to mention my wire is bare... why I have been thinking of some sort of comb structure.

Where I am most lacking is the understanding of the inductive geometries.

What do you think about a cylindrical form, with the wire zig zagged from top to bottom....

Like: VVVVVVVVVVVV

I can't even speculate as to the inductance of such beast... all of this may be overkill at 60 Hz. I really need to use the better portion of the 100 meter lengths to keep the power to surface area ratio ratio good.

I will fool around with 4 wire measurements on this meter... I had one of these when they first came out... the 4 wire measurements did not instill confidence.



  Reply With Quote
Old 30th January 2008, 03:49 AM   #4
SY is offline SY  United States
diyAudio Moderator
 
SY's Avatar
 
Join Date: Oct 2002
Location: Chicagoland
Blog Entries: 1
The simplest noninductive geometry can be visualized this way: bend the wire in half. Take the pointy end and attach it to the form. Now wind both leads simultaneously in the same direction. Or you can do an Ayrton-Perry winding, where you have one winding wound directly on top of the other, with opposite winding directions run in parallel.

This is sort of 1968 Ham Radio level stuff. I'm sure that John has a much nicer way to do it.

I'd use ceramic tubing, just for thermal conductivity.
__________________
"The way to deal with superstition is not to be polite to it, but to tackle it with all arms, and so rout it, cripple it, and make it forever infamous and ridiculous."- H. L. Mencken
  Reply With Quote
Old 30th January 2008, 04:06 AM   #5
diyAudio Member
 
Join Date: Mar 2007
Location: Canandaigua, NY USA
1968? More like 1896! The bare wire is a problem. I usually buy it with some modern insulation. Best bet might be to make up some separate modules on a notched form, then run them in series or parallel in the pail of oil. The inductance of any one will be minimal if you wind it the way Sy describes.

There's another cute way to do this, using the same technique as used in ten turn pots. If you pull one apart, you'll find the wire is wound in a very small coil around a piece of heavy enameled wire. Because the diameter of the coil is so small, the inductance is also small. the copper wire kills the inductance further. Now, the whole tiny coil is coiled up around the inside of the pot. Few turns equals less inductance, and it probably can be arranged to cancel out the contribution of the small diameter coil. You could wind around a piece of enameled wire, as long as you have a slight space between turns. Then, shellac it to keep everything in place. Coil it up a few inches in diameter and immerse it in the pail.

I have a fixture I find useful for all sorts of coil and resistor winding. It's nothing more than a drill chuck on an axle, with a crank on the other end. It turns in a big block of whatever on some scrap bearings. Great for making inductors, resistors, and springs.
  Reply With Quote
Old 30th January 2008, 04:18 AM   #6
poobah is offline poobah  United States
diyAudio Member
 
poobah's Avatar
 
Join Date: Nov 2005
Yeah I got that part... superposition of opposing amp-turns and all. I was questioning the twisting the wires after the fold had been made... I have naked wire coming.

The trick is is finding a fairly easy method to construct. I am thinking about folding the wire and THEN doing some sort of zig-zag. The cylinder I envision is just the shape of the coil... no real "pipe" involved... cylinder is attractive because I can use monster PVC and such for a vessel maybe?

And yes... I'm hoping to constrain the neutron toward an offering. He might be able to do a nifty SIM or model... might be sweating inductance irrelavant to the task?

I know he won't CHARGE me...

  Reply With Quote
Old 30th January 2008, 04:32 AM   #7
diyAudio Member
 
Join Date: Mar 2007
Location: Canandaigua, NY USA
PVC won't take a lot of heat, but hopefully the oil will solve that. For stability you don't want the wire to get too hot anyway. If you have a lathe, here's a trick I've used with Tesla coils- put a fine thread on the PVC pipe and wind in that. You could even do a double start thread so both wires would track together. No, don't try twisting the bare wire! My experience, and the vector impedance meter doesn't lie, is that only the folded method is worth much. It's good even if you wind around a big cylindrical form. Any attempt I've made at counter winding coils (that's how Julie Research did it commercially) really didn't reduce inductance that much. The proximity wasn't good enough. Flat windings (mica card resistors) are supposed to be pretty good, but again they don't come close to the bight method. OTOH, for low values, i.e., not much wire, none of this may matter at 60 Hz.
  Reply With Quote
Old 30th January 2008, 04:35 AM   #8
poobah is offline poobah  United States
diyAudio Member
 
poobah's Avatar
 
Join Date: Nov 2005
You're right Conrad...

Manganin was used for the legal ohm standard since 1901... not anymore though... they have gone with blue resistors now exclusively.

Part of the attraction of the zig zag is the fact that I can consolidate taps at one end of the coil support structure. And, I can do it with a "folded wire". I will need to tap the element... probably at ~log intervals... practicallity and power measurements over a linear scale.

Really... were it not for the doubt it would cast on the remainder of the design... I would just use the 101% eff. spec.... it is rather impressive after all.

In either event, my client is happy, having efficiency so high that it challenges measurement (mildly) is not a bad problem to have.
  Reply With Quote
Old 30th January 2008, 04:42 AM   #9
poobah is offline poobah  United States
diyAudio Member
 
poobah's Avatar
 
Join Date: Nov 2005
I was thinking about a threaded pipe.

And yes... temperature here will not go past 35-40 C. That is where the beauty of manganin fades. So material choices are pretty open. Mineral oil is baby butt compliant... hardens soft PVCs... does little to rigid PVCs.

The approach here is 2 fisted... stable wire in the first place. And then a good characterization of the R versus T.
  Reply With Quote
Old 30th January 2008, 06:46 AM   #10
Eva is offline Eva  Spain
diyAudio Member
 
Eva's Avatar
 
Join Date: Oct 2003
Location: Near the sea
Send a message via MSN to Eva
Soon I will have to face the task of doing precise efficiency measurements and tuning, as I'm being asked for 95% or better efficiency in a sine inverter at 2400W (including power supply!) I will only have a single power analyzer, so the obvious solution is automatic real-time switching of the analyzer between input and output. I have got more than enough 30A relays and PICs...

I wouldn't worry that much with precision load resistors in oil bath. Losses in the wiring and connections to the resistors may ruin the measurement anyway.
__________________
I use to feel like the small child in The Emperor's New Clothes tale
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
high precision high voltage supply nhuwar Planars & Exotics 13 12th February 2007 08:24 AM
Precision resistor source in Australia wintermute Parts 9 19th December 2004 07:58 PM
Spectrol, high precision, 10k attenuator. Peter Daniel Swap Meet 3 4th February 2004 05:15 PM
Precision and/or high performance SMD caps peranders Parts 21 17th April 2003 12:08 PM


New To Site? Need Help?

All times are GMT. The time now is 03:07 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2