Home |
Search |
Today's Posts |
|
#1
|
|||
|
|||
Toroidal core lamination properties.
A search on Google about GOSS brought a shirtload of mainly indecipherable info re GOSS transformer materials, and magnetic materials in general. The people at Tortech ( aust ), http://www.tortech.com.au/toranalysis.html say this "The magnetic core material for most power transformers Is made of grain-oriented, cold-rolled, 3% silicon steel that is coated and Insulated. That material has lower exciting current and core losses than regular steel. It also has relatively high saturation flux density with a high degree of squareness. Squareness is the ratio of residual flux density (remanence) to the maximum flux density (saturation), or Br/Bs. The oriented grain allows the steel to be operated at a higher saturation-flux density than non-oriented steel. The core material is annealed at high temperature in a dry hydrogen furnace to remove impurities and relieve the material stresses. Annealing also develops the desired magnetic properties, such as high squareness and low core loss. The steel strip Is coated with a chemical finish to ensure high resistance between laminations. Finally the annealed cores are varnish impregnated, cured. and painted or epoxy coated." There is reference to " most power transformers " being made from GOSS. But GOSS isn't universally used, its twice the price of the non oriented steels. There is no mention of the U figure. It is sales pitch. Most transformer companies like this don't make it easy for the layman to understand what is really available, and rely on the general impression that after several pages of boring transformer talk, a possible customer will conclude that they know a lot and place an order. Probably Tortech do use GOSS with a high U but we are left to assume they do. And at the end of the day we have to try some material and measure the VI input, and if its a low value, and losses are low and that is what is wanted, as well as low thd in audio trannies, we accept the material. We don't have to know all there is to be known about the core material production, but there is no reason to assume anything, and i base my position on the variability of core materials i have been sent from various companies in Oz over the last 8 years. Patrick Turner. |
#2
|
|||
|
|||
"Patrick Turner" = asinine, criminal ****wit The people at Tortech ( aust ), http://www.tortech.com.au/toranalysis.html say this ** Tortech is a small transformer winding business in Sydney. They buy lams and wound cores from local and foreign suppliers. The owner is Mike arkin - he supplied me with the 160VA wound core I use in my test. It is the same as ones he uses to make AC power transformers. The para quoted below does not even contain the word "toroidal " so has no relevance to the heading - as is usual for the Turneroid autistic . These pages all give the steel strip material used by toroidal mains transformer makers as being GOSS. http://www.airlinktransformers.com/t...al_toroid.html http://www.tabtronics.com/toroids.htm#core http://www.isomatic.co.uk/toroidalcores.htm http://avellindberg.com/transformers...ech_notes2.htm ................. Phil "The magnetic core material for most power transformers Is made of grain-oriented, cold-rolled, 3% silicon steel that is coated and Insulated. That material has lower exciting current and core losses than regular steel. It also has relatively high saturation flux density with a high degree of squareness. Squareness is the ratio of residual flux density (remanence) to the maximum flux density (saturation), or Br/Bs. The oriented grain allows the steel to be operated at a higher saturation-flux density than non-oriented steel. The core material is annealed at high temperature in a dry hydrogen furnace to remove impurities and relieve the material stresses. Annealing also develops the desired magnetic properties, such as high squareness and low core loss. The steel strip Is coated with a chemical finish to ensure high resistance between laminations. Finally the annealed cores are varnish impregnated, cured. and painted or epoxy coated." |
#3
|
|||
|
|||
Phil Allison wrote: "Patrick Turner" = asinine, criminal ****wit The people at Tortech ( aust ), http://www.tortech.com.au/toranalysis.html say this ** Tortech is a small transformer winding business in Sydney. They buy lams and wound cores from local and foreign suppliers. The owner is Mike arkin - he supplied me with the 160VA wound core I use in my test. It is the same as ones he uses to make AC power transformers. I know Mike Larkin. In 1995, he quoted me a huge figure for a PT, choke, and two OPTs. He said he was unable to do the OPTs in Oz because the finest wire he could use was 0.63 mm dia, and that he'd have to subcontract the OPTs to a company called NZ Coils P/L, which I have never been able to track down. I thanked him, and said I'd consider the deal for awhile. Since the material costs were about 20% of the huge quote, I figured i'd be better off winding all this stuff myself, because my hourly rates are far lower than his. It is impossible for any Oz transformer winder to make me anything at a price where I could then use in amps which I could then sell to compete with chinese imports and make a wage. Mike would understand this. It is most likely you have good GOSS material in the test core you used in your test. Its not hard to find out just what you got. Just measure the inductance at 50 Hz across a range of voltages until saturation occurs, and note the maximum L. Then work out what the Bmax is for where L is at a maximum. Probably you may find L max is where B = about 0.8T. Knowing what turns, Afe, and ML you got, you can work out what the U must be. My point I made previously was that since there is such a **HUGE** variation in the U and losses in E&I lams and Ccores I have purchased, it follows thet not all of what is claimed to be GOSS is identical. And I maintain that just because some GOSS is wound into a spiral, the U compared to an interleaved E&I tranny won't necessarily be 10 to 15 times what it is in the E&I lams. And I would maintain that if you had some "plain old transformer iron, ie, not GOSS, and you wound some into a toroid, there would be a considerable increase in the U compared to E&I, but just how much I don't know. The para quoted below does not even contain the word "toroidal " so has no relevance to the heading - as is usual for the Turneroid autistic . But it was from Tortech, who praise the virtues of their toroidal wares... These pages all give the steel strip material used by toroidal mains transformer makers as being GOSS. http://www.airlinktransformers.com/t...al_toroid.html http://www.tabtronics.com/toroids.htm#core http://www.isomatic.co.uk/toroidalcores.htm http://avellindberg.com/transformers...ech_notes2.htm ................ Phil Thankyou for the references. From one they say "Since toroidal cores are constructed of a continuously wound ribbon, there is virtually no air gap. The windings are evenly wrapped over the entire core allowing the transformer to operate at a higher flux density than in standard transformers. Toroidal transformers can operate at 1.6 to 1.8 Tesla (16,000 to 18,000 Gauss) while EI cores are limited to 1.2 to 1.4 Tesla (12,000 to 14,000 Gauss)." I have never seen a toroidal tranny set up to run at 1.8T ***ALL*** the several toroidals from Jaycar were set up to run at no more than 1.25T, and all were disgustingly noisy and unacceptable in hi-fi amps. Two 800 VA toroidal transformers made by Harbuch were also unacceptably noisy, and also set up to run at 1.25T, and the only way I could shut the buggers up was to remove the secondaries by winding the wire onto a notched dowel shuttle by hand, and add 33% more turns to the primary, and then rewind the primary, all with a shuttle. It took days. But I got there, and the trannies finished up acceptably quiet for use in tube amps with a main power going to a rectified plate plate supply, despite the fact that the re-wound tranies were not vacuum varnished, but empployed the same insulation technique of winding tight strips of polyester on which is then further self tightening when layers of wire are pulled tight ove them. One amp has been in constant use since 1996, and the tranny runs quiet. Needless to say, the core losses are negligible, but I measured it all, since i assume nothing. Since getting toroidal OPTs properly wound with all the taps and multiple windings i wanted and with epoxy resin impregnant vaccuum impregnated is prohibitively expensive and unlikely to even be available, I was forced to look around for decent low loss E&I, and I came across Sankey who make the GOSS E&I that measures U = 17,000. I have to buy $400 worth each time, since their minimum order is 50 Kg. I had previously bought some british made imported GOSS E&I, I think it went about 7,000, not too bad.... Plitron wind some splendid toroidal transformers. They do cost quite a bit though. Patrick Turner. |
#4
|
|||
|
|||
"Patrick Turner" = asinine, criminal ****wit Phil Allison It is most likely you have good GOSS material in the test core you used in your test. ** It is a certainty since that is the only material used. And I maintain that just because some GOSS is wound into a spiral, the U compared to an interleaved E&I tranny won't necessarily be 10 to 15 times what it is in the E&I lams. ** Two counter examples: #1. The I mag @ 240 volts for a 300 VA mains toroidal ( from Harbuch) on the bench is 11mA rms. At 120 volts, I mag falls to 1.7 mA, becomes sine wave and has a 12 degree phase angle. The load R is effectively 70 kohms. For this case, L computes to be over 1000 H using: L = R / ( 2.pi.50. tan 12d ) #2. The I mag @ 240 volts for a 160 VA mains toroidal ( from Jaycar) on the bench is 7 mA rms. At 120 volts, the I mag falls to 1.6 mA, becomes near sine wave and has a 15 degree phase angle. The load R is effectively 75 kohms. L computes to be nearly 900 H. The para quoted below does not even contain the word "toroidal " so has no relevance to the heading - as is usual for the Turneroid autistic . But it was from Tortech, who praise the virtues of their toroidal wares... ** More autistic Turneroid excrement - what a ****ing pong !!! ............... Phil |
#5
|
|||
|
|||
Phil Allison wrote: "Patrick Turner" = asinine, criminal ****wit Phil Allison It is most likely you have good GOSS material in the test core you used in your test. ** It is a certainty since that is the only material used. And I maintain that just because some GOSS is wound into a spiral, the U compared to an interleaved E&I tranny won't necessarily be 10 to 15 times what it is in the E&I lams. ** Two counter examples: #1. The I mag @ 240 volts for a 300 VA mains toroidal ( from Harbuch) on the bench is 11mA rms. At 120 volts, I mag falls to 1.7 mA, becomes sine wave and has a 12 degree phase angle. The load R is effectively 70 kohms. For this case, L computes to be over 1000 H using: L = R / ( 2.pi.50. tan 12d ) #2. The I mag @ 240 volts for a 160 VA mains toroidal ( from Jaycar) on the bench is 7 mA rms. At 120 volts, the I mag falls to 1.6 mA, becomes near sine wave and has a 15 degree phase angle. The load R is effectively 75 kohms. L computes to be nearly 900 H. You are quite right with all these calcs I get quite high enough inductance figures with the GOSS E&I material I use from Sankey. For an iron cored inductor, with no unbalanced DC flow in only one direction, as in an SE amp, or with balanced DC, as in the case of a PP amp, or a mains tranny is 1.26 x Np squared x Afe x U Lp = --------------------------------- 1,000,000,000 x ML Where Lp = primary inductance in Henrys, 1.26 is a constant, Np = primary turns, Afe is the cross sectional area of the core in square mm, U is the mu of the iron at low voltage levels, 1,000,000,000 is a constant, ML is the magnetic path length around the core in mm. Now its obvious from your measurements that the L was more at 120v than 240v input, which tells you the U of the material has declined with increasing voltage, ie increasing Bmax. You know all the items in the above formula except U. What is the U for the 240v and 120v conditions? If the current at 240v is badly distorted, the calculation of L at the higher Vin is questionable. But max U may be well below saturation onset, and maybe at 180v input. So what maximum U are you getting? I try to run my trannies at about 0.85T, and U is at its highest, losses are negligible, and even with crummy NOSS E&I cores the outcome is acceptable because the item is quiet and cool. The para quoted below does not even contain the word "toroidal " so has no relevance to the heading - as is usual for the Turneroid autistic . But it was from Tortech, who praise the virtues of their toroidal wares... ** More autistic Turneroid excrement - what a ****ing pong !!! I didn't think a pong could copulate, but anything is possible in your imagination. Patrick Turner. .............. Phil |
#6
|
|||
|
|||
"Patrick Turner" = asinine, criminal ****wit Phil Allison And I maintain that just because some GOSS is wound into a spiral, the U compared to an interleaved E&I tranny won't necessarily be 10 to 15 times what it is in the E&I lams. ** Two counter examples: #1. The I mag @ 240 volts for a 300 VA mains toroidal ( from Harbuch) on the bench is 11mA rms. At 120 volts, I mag falls to 1.7 mA, becomes sine wave and has a 12 degree phase angle. The load R is effectively 70 kohms. For this case, L computes to be over 1000 H using: L = R / ( 2.pi.50. tan 12d ) #2. The I mag @ 240 volts for a 160 VA mains toroidal ( from Jaycar) on the bench is 7 mA rms. At 120 volts, the I mag falls to 1.6 mA, becomes near sine wave and has a 15 degree phase angle. The load R is effectively 75 kohms. L computes to be nearly 900 H. You are quite right with all these calcs ** Now you do the same for E -core tannies of similar ratings - your own or commercial ones. Lets compares the I mag and effective L primary figures. Lets see if the toroidals win by 10 to 15 times. ............... Phil |
Reply |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
KISS 117 by Andre Jute | Vacuum Tubes | |||
Transformer calculations... | Vacuum Tubes | |||
Toroidal Core Info | Vacuum Tubes | |||
Stupid transformer question | Vacuum Tubes | |||
Help: How to wind a toroidal output transformer for push-pullapplications? | Vacuum Tubes |