radiomuseum.org
Please click your language flag. Bitte Sprachflagge klicken.
 

Trafo Bestimmung

Moderators:
Ernst Erb Jürgen Stichling Bernhard Nagel 
 
Please click the blue info button to read more about this page.
Forum » Technique, Repair, Restoration, Home construction ** » Home construction and general radio technique » Trafo Bestimmung
           
Gerhard Bischof
Gerhard Bischof
 
A  Articles: 57
Schem.: 61
Pict.: 67
11.Jul.09 19:32

Count of Thanks: 12
Reply  |  You aren't logged in. (Guest)   1

Hallo und guten Tag

 

Ich erlaube mir, meine Erfahrungen bei der Bestimmung der wichtigsten Kenngrößen

eines unbekannten Trafos als PDF-Dokument zur Verfügung zu stellen.

 

mit freundlichen Grüßen

Gerhard Bischof

 

EDIT: ich muß die pdf-datei in 3 teile schneiden da sie über 224 KB hat.....................

p.s. es wäre leichter und weniger zeitaufwendig für autoren

wenn pdf -Dateien größer als 200 kByte zugelassen würden.

Attachments:

This article was edited 11.Jul.09 20:36 by Gerhard Bischof .

Joe Sousa
Joe Sousa
Editor
USA  Articles: 664
Schem.: 214
Pict.: 434
12.Jul.09 01:13

Count of Thanks: 8
Reply  |  You aren't logged in. (Guest)   2

Hallo Gerhard - mit der Entschuldigung für die Google-Übersetzung.

Vielen Dank für Ihre Buchung Messungen dieser Transformator in einer sehr klaren Form. Ich habe Google Translate zu lesen, Ihre Post-und PDF-Dateien.

Der Transformator sieht wie es aus dem Netzteil der Mikrowelle. Die dicken, isolierter Draht, mit nur zwei sich zwischen den bobins ist die Faserwickelmaschinen für die mangnetron. In der Anwendung der overn dicken Draht bobin ist der erste und der dünnen Draht bobin ist die sekundäre mit einem der Ausgang HV Drähte an den Kern.

Normalerweise werden diese Transformatoren haben eine magnetische Kern-Shunt zwischen der primären und sekundären bobins. Dies erhöht die Dichtheit Shunt Induktivität, mit dessen Hilfe die Leitung Winkel des halben Einweggleichrichter, und somit Leistungsfaktoren in mehr als 90%. Ich habe einen Ofen unter Last eine Weile zurück und haben diese Macht-Faktor.

Die Fotos zeigen nicht eindeutig, wenn Ihr die Messungen durchgeführt wurden, ohne die magnetischen Kern-Shunt. Die Anwesenheit der Shunt würde wesentlich niedriger die Quote macht.

Wenn ich war auf der Suche in den Betrieb von 60Hz Transformatoren vor kurzem habe ich festgestellt, dass die Menge von Stahl wird in der Regel Unternehmen für den Betrieb direkt am Rand der Sättigung. Erhöhung der Eingangsspannung von 10% doppelt so hoch wie die aktuelle Eingabe Magnetisierung. Die seltsame Sache ist, dass ich gefunden habe, dieses Merkmal in 5W Wand Transformatoren und 700W Mikrowelle Transformatoren.

Meine Frage ist: War der magnetischen Kern-Shunt in der Transformator während Ihres Messungen?

Grüße,

-Joe

 


Hello Gerhard,

Thanks for posting your measurements of this transformer in a very clear form. I used google translate to read your post and pdf files.

The transformer looks like it came from the power supply of a microwave oven. The thick insulated wire with just two turns between the bobins is the filament winding for the mangnetron. In the overn application the thick wire bobin is the primary and the thin wire bobin is the secondary with one of the output HV wires connected to the core.

Usually these transformers have a magnetic core shunt between the primary and secondary bobins. This shunt increases leakage inductance, which helps increase the conduction angle of the half wave rectifier, and thus obtain power factors in excess of 90%. I measured an oven under load a while back and got this power factor.

The photos don't show clearly if your measurements were done without the magnetic core shunt. The presence of the shunt would substantially lower the turns ratio.

When I was looking into the operation of 60Hz transformers recently, I noticed that the amount of steel is usualy sized to operate right at the edge of saturation. Increasing input voltage by 10% can double the input magnetization current. The odd thing is that I have found this characteristic in 5W wall transformers and in 700W microwave oven transformers.

My question is: was the magnetic core shunt in the transformer during your measurements?

Regards,

-Joe

Gerhard Bischof
Gerhard Bischof
 
A  Articles: 57
Schem.: 61
Pict.: 67
12.Jul.09 13:31

Count of Thanks: 8
Reply  |  You aren't logged in. (Guest)   3

hello joe

thanks for your info.

i did not know that this is a microwave trafo.

the trafos i dealed with all reduce the voltage......;o)

you made the right conclusions.

indeed there was a solder-point from the "photo-primary" to the iron core.

i also thought the thick orange wire between the two bobbins is a construction detail of the coil-winder and trafo assembler to get space between the bobbins. i got the trafo as it is.

also there are a bunch of little I laminates between the bobbins to prevent vibrating i think. ( hole is too big for the liquid adhesive...)

i never saw welded iron core-laminates before. i thouhgt this is because there are no mounting holes in

the E laminates. maybe a mal-production of laminates that was bought cheaply and used for trafo production by the campany anyway.... is this conclusion ok??

i´m only at the beginning of my interrest in trafo and output transformer design.

i got a micafill trafo coil winding machine from switzerland and now i try out everything i get to read or to studiy. bit at first i use "scrap-trafos" for getting the right skill for this work. i bought a lot new cores, bobbins an copper wire but they are too expensive for experiments....

thank you so much for your comments

 

best regards

gerhard

 

 

 

 

 

Joe Sousa
Joe Sousa
Editor
USA  Articles: 664
Schem.: 214
Pict.: 434
14.Jul.09 06:54

Count of Thanks: 12
Reply  |  You aren't logged in. (Guest)   4

Hello Gerhard,

Transformers and inductors are very interesting devices. Their characteristics are almost never ideal. Every application requires careful consideration. The deviation from the ideal may be in the linear characteristics, reactive (capacitive or inductive), and even non-linear behaviour.

Up to the 1950's most high power low impedance low frequency amplification and control was done with magnetic amplifiers. Then transistors took over this job. Magnetic amplifiers worked on the general idea that a small amount of low frequency power, let's say DC power, could control the saturation of a reactor (=cored inductor) and greatly affect how much AC power could pass through this reactor. One simple application was for low loss light dimmers. One peculiar use of a magnetic amplifier was in the GE colorama console radio from the late 1930's. A few mA from a control triode driven by the AGC line could bring a reactor in and out of saturation. The reactor was in series with red and green light bulbs that would turn green when the station was tuned in.

http://www.angelfire.com/electronic2/radiosean/tv/GE_E91/GE_E91.htm

http://www.angelfire.com/electronic2/radiosean/tv/GE_E91/Riders_GE_7-25_28.pdf

OK, I am really glad you have taken an interest in inductors, I am just trying to add to it.

---------------Now, about your transformer:

One of the things that would be good to measure in your transformer would be leakage inductance, which is defined as the inductance that is not common to primary and secondary. There is a primary leakage inductance and secondary leakage inductance.

One easy way to measure the leakage inductances is to short one winding while measuring how much inductance is left on the other. You will get as many values are there are windings.

Microwave oven transformers are designed with a lot of leakage inductance, as provided by the magnetic shunt. You may like to measure how much leakage inductance you get with, and without the shunt. With the shunt, the leakage inductance will be quite high, perhaps a third of the open circuit inductance.

You should also measure the open circuit inductances for comparison.

One advantage of using the transformer in the direction you specified, as a step-down transformer, is that the transformer will be very linear.

I made a few casual measurements on a microwave transformer in the step up configuration and got something like 50W of unloaded dissipation, most of it in the core saturation. I would be surprised if this same transformer had an iddle dissipation of 5W in the step-down configuration.

I look forward to any new measurements you may make.

Regards,

-Joe


Hallo Gerhard, - wieder, mit Entschuldigung für die Google-Übersetzung.

Transformatoren und Drosseln sind sehr interessante Geräte. Ihre Merkmale sind fast nie ideal. Jede Anwendung erfordert eine sorgfältige Prüfung. Die Abweichung von der idealen kann in der linearen Merkmale, reaktive (kapazitiv oder induktiv), und auch nicht-lineare Verhalten.

Bis zu den 1950er Jahren die meisten Hochleistungs-niederohmigen niedrigen Frequenz Verstärkung und Kontrolle wurde mit magnetischen Verstärker. Transistoren Dann übernahm diese Aufgabe. Magnetische Verstärker an die allgemeine Vorstellung, dass eine kleine Menge von niederfrequenten Macht, sagen wir mal, DC-Netzteil, könnte die Sättigung eines Reaktors (= Kerngehäuse Induktion) und sehr viel Einfluss darauf, wie das Stromnetz könnte durch dieses Reaktors. Eine einfache Anwendung wurde für geringen Verlusten Licht Dimmer. Eine besondere Verwendung eines magnetischen Verstärker wurde in der GE Colorama Konsole Radio aus den späten 1930er Jahren. Ein paar mA aus einer Triode von der AGC-Linie könnte einen Reaktor in die und aus der Sättigung. Der Reaktor wurde in Reihe mit dem roten und grünen Lampen, die grün, wenn die abgestimmt wurde in.

http://www.angelfire.com/electronic2/radiosean/tv/GE_E91/GE_E91.htm

http://www.angelfire.com/electronic2/radiosean/tv/GE_E91/Riders_GE_7-25_28.pdf


OK, ich bin wirklich froh, Sie haben ein Interesse an der Drossel, ich bin nur versuchen, in den es.

--------------- Nun, zu Ihrer Transformator:

Eines der Dinge, wäre gut für die Messung in Ihrem Transformator wäre Leckage Induktivität, die definiert ist als die Induktivität, die nicht in der Primar-und Sekundarstufe. Es gibt eine primäre und sekundäre Leckage Induktivität Leckage Induktivität.

Eine einfache Methode zur Messung der Leckage-Induktivitäten ist kurz Wicklung während der Messung, wie viel Induktivität ist links auf der anderen Seite. Sie erhalten so viele Werte gibt es Windungen.

Mikrowelle Transformatoren sind mit einer Menge von Leckagen Induktivität, wie es von der magnetischen Shunt. Sie können, wie um zu messen, wie viel Leckage Induktivität Sie mit und ohne Shunt. Mit dem Shunt, der Leckage-Induktivität wird zwar recht hoch sein, vielleicht ein Drittel der Leerlauf Induktivität.

Sie sollten auch Maßnahmen der offenen Schaltung Induktivitäten für den Vergleich.

Ein Vorteil der Verwendung der Transformator in die Richtung, in die Sie bestimmt ist, als Step-down-Transformator, ist, dass der Transformator wird sehr linear.

Ich habe ein paar lässige Messungen auf einer Mikrowelle Transformator in der Schritt-Konfiguration und bekam so etwas wie der unbelasteten 50W Verlustleistung, die meisten davon in den Kern Sättigung. Ich würde mich wundern, wenn dieser Transformator hatte eine Verlustleistung von 5W ITTE in der Step-down-Konfiguration.

Ich freue mich auf jede neue Messungen können Sie.

Grüße,

-Joe

 

Gerhard Bischof
Gerhard Bischof
 
A  Articles: 57
Schem.: 61
Pict.: 67
15.Jul.09 15:09

Count of Thanks: 8
Reply  |  You aren't logged in. (Guest)   5

hello joe

please would you tell me:

i took 2 pieces of this trafo, and reassembled one new trafo with two peaces of the "photo-primary".

( i.e. 2 x 2264 windings approx.) to get a galvanic-separated trafo. i welded it together and got

primary 225 VAC and secondary 230 VAC with no load.  it seemed to be "perfect".

but with a load of 100 watts ( lamp) the secondary voltage decreased to 86 VAC. my conclusion:

due to the welding there is a disturbing in the flux density in the iron. maybe lots of eddy-loss??

is this due to welding or is it because a certain kind of iron for this trafos??

or is there a leakage of flux in the core?? if yes ---  why?

thanks for info in advance

best regards

gerhard

Joe Sousa
Joe Sousa
Editor
USA  Articles: 664
Schem.: 214
Pict.: 434
15.Jul.09 20:21

Count of Thanks: 6
Reply  |  You aren't logged in. (Guest)   6

Hello Gerhard,

My first guess would be that there is too much resistance in the windings. I don't think that there are any significant saturation losses because the new primary inductance is much higher than the original design, so there is much less net magnetization flux. A higher inductance wins in reducing flux because it increases by the square of the ampere-turns. So more turns reduces the flux for a given AC drive voltage.

Eddy current loss should be no worse than for the original design. Welding can only increase, not decrease eddy losses. The welding seams are placed to have a minimal effect on eddy losses.

I think you gave me enough data for an estimate of the winding resistance.

The resistance of the 100W 240V light bulb can be estimated from Rustika 60W Bulb resistance

(this file was originally used in Rustika lightbulb FM measurements)

A 100W bulb draws 100W/240V=416mA R100W=528_Ohms. The 60W Rustika Bulb resistance profile suggest that at 86V, the  resistance for the 100W bulb will be 358_Ohms and the current will be 240mA. This estimate involves the assumption that a gas-filled 100W bulb will have the same resistance-vs-voltage profile as the 60W Rustika bulb which is in vacuum. This assumption is not 100% true, but it is close enough for a first estimate.

A simple extrapolation for a 358_Ohm bulb resistance causing a 86V drop, gives a total transformer equivalent droping resistance of 641_Ohms. The windings are identical, so around 300_Ohms is expected at each winding.

This resistance estimation assumed that the leakage inductance ( leakage inductance is in the transformer, but circuit wise, seems to be in series with it) is small enough at 50Hz to be neglible.

So the 300_Ohm resistance should be though of more as an impedance.

If the winding resistance were negligible, and all the transformer drop could be blamed on leakage inductance, then the magnitude of the inductive impedance would be 933_Ohms, or 2.9H of total leakage inductance; 1.45H on each winding. This seems high with the magnetic shunt removed.

My vote is that the impedance is mostly resistive around 300 Ohms per winding. You can let us know with simple resistance measurements on primary and secondary.

Regards,

-Joe

 

Gerhard Bischof
Gerhard Bischof
 
A  Articles: 57
Schem.: 61
Pict.: 67
15.Jul.09 22:57

Count of Thanks: 4
Reply  |  You aren't logged in. (Guest)   7

hello joe

thank you so much for your informative comments. the more i read your text the more it is going to get clear in my mind........

the resistances i measured primary and secondary were 106 and 108 Ohms respectively.(fluke multimeter).

please excuse but i cannot follow your explanation about the 100 W lamp at 86 V.

in your Rustika_Filament_DC_drain.MCD picture, which occures as an inverse parabola to me ( the slope must be the inverse of resistance "ohm"  named "siemens"..... ) i can read at 86 Volt

140mA approx.  yielding 614 ohms. whats wrong in my interpretation?

( maybe your estimates are for a drop of 86 V?? the voltage was at 86 V !!!!)

looking at your mathcad-sheets i see that you are a physicist with huge  experiances in electrics. seems that i´m an electic rooky. i appreciate that you are talking to me.....

i studied mathematics 18 years ago and now i´m trying to "apply" this sience---;o)

joe, i have another question concerning the flux density B in the trafo main-equation:

some producer of iron laminates give data for the core-laminates and name a flux-desity value

e.g. 1,4 Tesla. is this the maximum for this sort of iron or is this the proper "working-flux-density" ??

what value of flux-density should i take if i want to create a power trafo for example?

i read in some books that values from 0,5 - 1,7 Tesla can be taken. how do i decide which value??

would be very nice if had an info before my brain is breaking...;o)

 

thanks a lot

best regards

gerhard

Joe Sousa
Joe Sousa
Editor
USA  Articles: 664
Schem.: 214
Pict.: 434
16.Jul.09 04:26

Count of Thanks: 7
Reply  |  You aren't logged in. (Guest)   8

Hello Gerhard,

Your resistance measurement values of 100_Ohms per winding are well within the expected range. It appears that there is some significant leakage inductance taking up most of the remaining drop. The other source of error that we know about, is the difference in resistance profile for a 60W Vacuum bulb verses a 100W bulb with a double-helix of tungsten in inert gas. A lower bulb resistance would give a lower transformer resistance estimate.

I just checked the calculation for the bulb resistance by looking at the Voltage vs Current graph, and it seems correct. However, a major clarification is in order here. The resistance plot for the 60W Rustika represents the slope of the V-vs-I plot. This slope plot is the correct one to use for small signals on top of the high power DC or AC voltage.

But the correct plot to use for the transformer plot is the IvsV plot. The resistance values I gave are direct V/I ratios, they are not slopes. The calculations I just made using the I/V plot still came to a current draw of 241mA at 86V, for the 100W bulb estimation.

Sorry for the quick earlier calculation, it led to legitimate confusion.

If you have an AC ampmeter, then you could know exactly how much current the bulb is drawing. The current wave should be a clean sinewave, so that even an ampmeter that is not true-rms will be pretty accurate.

-------------------------------------

About MathCAD: It produces impressive printouts, but that does not imply great knowledge or training.  Some of the plots were from my measurements, and the plot at the end, making an equivalency to temperature, was based on a table I found at a thermometer vendor.  My main experience is all in electronics engineering as a profession and as a hobby. Advanced Physics and Mathematics lie at the boundaries of what I know.

My main interest in the forum is learning and sharing, in equal measures. Besides, I don't think I have ever been able to share without learning, or to learn without sharing. Often I learn more with people who are also trying to figure things out, like you and me.

-------------------------------------

About flux density:

I don't have direct experience measuring the flux density of the transformers that I have, otherwise, measured. The Tesla values from the manufacturer are a good start in choosing the proper density.

The choice of the flux density is dictated by how much saturation are you willing to tolerate. As I found when I measured a transformer like yours, and the small transformer in the other thread, it appears that innexpensive transformers are designed pretty close to the edge of allowable saturation. Sometimes a design is so close, that it is rated for 60Hz, but not 50Hz, because the saturation losses go up disproportionately with reduced frequency. The cliff effect of saturation reveals itself. I would not be surprised if transformer losses doubled as a transformer went from 60Hz to 50Hz.

One further aspect about the particular design of the microwave transvormer is that they are never operated without a load. So the beneficial capacitive loading of a conventinal microwave transformer is taken into account in the design of the transformer, and may allow for a smaller core than would be reasonable without the capacitive loading.

One way to make a good isolation transformer from your transformer, but with lower saturation would be to use the thick wire for the primary and secondary, and increase the number of turns for each winding by perhaps 30%. One experiment that you could conduct, would be drive the original thick wire primary of your transformer with a variable transformer and see at what primary voltage the current starts to increase rapidly. If this voltage is 180VAC, then you add enough turns to increase the linear inductance such that the net core flux stayed at the 180VAC level when you apply 240VAC to the augmented winding. This just gives the concept. You can use your formulas to figure out the number of additional turns that would be required.

When you apply voltage to the original thick winding, be aware that about 2kV would be produced at the thin winding. This is not an issue if this winding has been removed.

So my basic criterion would be to stay away from saturation by direct measurement and sensible use of simple formulas because you don't have specifications for the core material.

Regards,

-Joe

Gerhard Bischof
Gerhard Bischof
 
A  Articles: 57
Schem.: 61
Pict.: 67
18.Jul.09 11:13

Count of Thanks: 6
Reply  |  You aren't logged in. (Guest)   9

hello joe

here i present  the last measurements (all secondary) under load before disassemling this trafo:

(the "most-first" one was with a lamp with unknown wattage ( i thought it was 100 Watts but wasn´t!!)

primary : 230VAC

1) lamp, 40 W on secondary ----> 94 VAC at 127 mA

2) lamp, 60 W on secondary ----> 51 VAC at 138 mA

3) lamp, 100W on seconsary ---> 14 VAC at 145 mA   the obvious voltage drop is really remarkable.

Noe i´ll try this core with the "photo-secondary" and will post my results.

greetings from austria

gerhard

 

 

  
rmXorg