The AM Forum
March 29, 2024, 08:51:28 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: Power transformer temperature during normal operation  (Read 10961 times)
0 Members and 1 Guest are viewing this topic.
AB2EZ
Member

Offline Offline

Posts: 1722


"Season's Greetings" looks okay to me...


« on: July 26, 2014, 12:40:03 PM »

Another thread raises questions regarding a transformer that is getting warm or hot.

Having never thought about how warm or hot one might expect a transformer in a piece of vintage equipment to get (what is normal/as designed and what is abnormal)...  I found the following reference on the web

http://ecmweb.com/content/basics-transformers-part-2

I my opinion it provides an explanation that is simple and concise... much more so than many of the other references I found.

Note:

As I interpret this:

If the insulation of the windings is rated for continuous operation (with a long expected life) at 105C (which is apparently not a particularly high rating among the various ratings that are now standardized), then one might expect the temperature of the core of a power transformer to rise to around 95C when the transformer is delivering its maximum rated power. I.e. this allows 10C for "hot spots".

Depending upon the thermal resistance between the core and the case, and how good the air circulation around the case is...one would expect the case temperature of the transformer to be lower than the core temperature... but still well above ambient temperature.

If the case temperature is (for example) 67.5C (i.e. half way between a 40C ambient temperature and a 95C core temperature)... then that corresponds to 153.5 degrees Fahrenheit.

So, perhaps a transformer that feels "warm" or "hot" to the touch (after running for a while at its intended power delivery) is not necessarily an indication of a problem.

Does anyone have any better information about how warm or hot one might expect the transformer to get during normal operation in a typical vintage Ranger or Valiant, etc. ?

Stu
Logged

Stewart ("Stu") Personick. Pictured: (from The New Yorker) "Season's Greetings" looks OK to me. Let's run it by the legal department
W3RSW
Contributing
Member
*
Offline Offline

Posts: 3308


Rick & "Roosevelt"


« Reply #1 on: July 26, 2014, 04:59:39 PM »

Subjective and haven't really measured temps; My plate trannys in HB equip. all run cold or coolish.  My filament (only) trans. run warm. All the multipurpose trans. In commercial equip. Seem to run pretty warm to the touch, some hot even after recapping, etc  particularly in tube receivers.  -I think Hallicrafters mostly since they seem to have about 2/3 the core of say an equivalent Hammarlund .

The LV in the 32v runs pretty warm. The plate runs same temp. as ambient inside the cab. from itself and nearby tubes.  The trans. In the Citation II AF amp runs pretty warm, some heat from the nearby KT88's, much warmer than the same, adjacently exposed AF outputs.

After awhile you get a feel for what's normal.  Reading your reference should ease a lot of our minds about what is normal.  Tnx.
Logged

RICK  *W3RSW*
AB2EZ
Member

Offline Offline

Posts: 1722


"Season's Greetings" looks okay to me...


« Reply #2 on: July 26, 2014, 06:28:09 PM »

Rick

Thank you for your inputs on this. I hope that we get more.

The transformer in my Ranger seems to run pretty warm/hot when I run the Ranger using its internal modulator. I've solid-stated the HV and LV power supplies (two rectifier tubes removed from their sockets)... but I'm using 6550's instead of the stock modulator tubes. I removed the 20k ohm wire wound bleeder/voltage divider resistor across the HV B+. I'm using a 100k ohm bleeder resistor across the HV B+, and a solid state, voltage regulated supply for the modulator tube screens. I replaced the audio driver transformer with a solid state phase splitter that includes a pair of source followers to drive the modulator tubes. This circuit consumes about 10 watts of power. I adjust the AC voltage to 115 volts. So, on balance, I think I have about the same load on the transformer as in a stock configuration.

Recently I modified the Ranger's 9-pin socket, so that when I remove the standard jumper plug, and plug in my external modulator, all of the Ranger's audio tubes (including the 6550's) have their filaments disconnected. This also turns off the power to the solid state audio phase splitter and the associated source followers.

That seems (qualitatively) to have allowed the Ranger's transformer to run much cooler.

It appears that providing filament power is a big factor in causing vintage equipment transformers to run warm/hot.

Stu
Logged

Stewart ("Stu") Personick. Pictured: (from The New Yorker) "Season's Greetings" looks OK to me. Let's run it by the legal department
W3RSW
Contributing
Member
*
Offline Offline

Posts: 3308


Rick & "Roosevelt"


« Reply #3 on: July 26, 2014, 09:40:23 PM »

Not to mention the stock Rangers horrible ventilation. I have a 4in. DC  Muffin fan on a wide portable and shallow wooden cigar box top, felt rimmed that covers about 2/3 the area of the Ranger's top. Pulls air out and blows upwards. Way cooler for everything inside. I move it to other needy cabs sometimes.  It's a 12v fan and the 6.3 v. / half wave supply small xformer provides about 8 volts for quieter running.

Even that little fan xformer gets warm with hardly any load.  Grin
Nature of inefficiency always generating heat in a closed system. " Hain't got no fins."
Logged

RICK  *W3RSW*
WQ9E
Member

Offline Offline

Posts: 3282



« Reply #4 on: July 27, 2014, 07:39:56 AM »

Drake 4 series receivers have very small and hot transformers.  I measured a few with my Fluke temperature meter but I can't remember their operating temperature.  A lot of the Halli receiver transformers do operate at high temperature also.

The Johnson Valiant operates at a very reasonable temperature compared to the Ranger which is a little toaster oven, a lot of tubes packed into a small space.  The Viking 500 is another heat generator even with its external power supply/modulator and the little final cooling fan doesn't do a lot.  The top of the 500 used to be my Siamese cat's favorite winter nap spot until one time the delay relay didn't and the arc gap fired, she didn't like waking up to a little lightning storm just beneath her bed Smiley  An added capacitor took care of the missing mechanical delay.

More modern stuff can be a problem also like my Yaesu FT-980 which was my first new rig back in 1983.  In receive only the very noisy cooling fan would come on after about 10 minutes and by that point the sink was far too hot to comfortably touch.  My guess is that the heat sink was greatly undersized, it is similar to the heat sink that was used on the FT-One which had a switching supply instead of the FT-980's linear regulator.  After about a year of this I hooked up an Astron external supply to provide the 12 volt bus and the internal supply supplied the 24.5 volt finals.  Operating this way it stayed cool even under contest conditions.   A few years ago I picked up a mint FT-One and its heat sink only gets slightly warm.
Logged

Rodger WQ9E
w1vtp
Member

Offline Offline

Posts: 2638



« Reply #5 on: July 27, 2014, 12:56:53 PM »

Boy, this thread brings back some memories.  An old timer that used to help me out with technical things said, "if you can hold your hand on a transformer, even though it very hot, it's OK."  If you can't hold your hand on it then you should be looking for a problem.

Not very scientific but it's served me well over the decades.

Al
Logged
W3RSW
Contributing
Member
*
Offline Offline

Posts: 3308


Rick & "Roosevelt"


« Reply #6 on: July 27, 2014, 01:32:31 PM »

So Al, that's how you earned the name "HOT HANDS."
We've always wondered.  Grin

-A time honored tradition; I understand that carrier pilots only get an aircom
call after fluffing a landing or some other mess-up.
Think I saw on a TV documentary where one pilot inadvertently did a fuel dump on landing what with all the other confusion going on.  His radio call forever more was "Dumpy" or something like that.
Logged

RICK  *W3RSW*
K4RT
Contributing
Member
*
Offline Offline

Posts: 520



« Reply #7 on: July 27, 2014, 05:35:54 PM »

Thanks for posting the link.

Were there any commercial CCS tube transmitters that monitored transformer core temps?  If so, how was it done?

With a potted ham radio transformer that runs hot but otherwise appears to be operating normally, I would think that adding a cooling fan to move air around the enclosure might be worthwhile to help draw some of the heat away.
Logged
w4bfs
W4 Beans For Supper
Contributing
Member
*
Offline Offline

Posts: 1433


more inpoot often yields more outpoot


« Reply #8 on: July 29, 2014, 05:04:43 AM »

hi Stu ... as to longevity, I remember an old rule of 10 for transformers and motors that stated that a 10 degree reduction in operating temperature can yield a doubling of operating life, or something like that.

one thing I have found that helps to cool off hot transformers is to solid state the rectifiers and use the freed up rectifier fil winding to buck the primary winding by a few volts .... this helps the old standard 110V transformers in the now 120V world ...73
Logged

Beefus

O would some power the gift give us
to see ourselves as others see us.
It would from many blunders free us.         Robert Burns
AB2EZ
Member

Offline Offline

Posts: 1722


"Season's Greetings" looks okay to me...


« Reply #9 on: July 29, 2014, 09:33:02 AM »

Beefus
Brad
Rick
Al
Rodger

Thanks for the comments.

I agree that reducing the average operating temperature of the components inside a piece of vintage equipment is a good thing to do. It is also a good idea to keep all DC supply voltages, applied to components, at or somewhat below their nominal (as originally designed) values.

I use a very quiet 3 inch muffin fan on top of my Ranger's cabinet (above the transformer). I adjust the AC line voltage to be 115V.

I haven't done this: but another reversible modification would be to use an auxiliary filament transformer to off-load the 6.3V and/or 12.6V filaments from the main transformer. Modern toroidal filament transformers are compact, readily available, and inexpensive. It would be easy to install a transformer like that on a bracket inside a vintage transmitter or receiver.

A relatively concise and simple summary of the effect of operating temperature on the expected lifetime of a component can be found here:

http://www.edn.com/electronics-news/4379372/The-Effect-of-Temperature-on-Failure-Rate

Note: the very last equation in the article has a typo in it. The "=" sign on the right side of the equation should be a "x" sign.

Summarizing what is said in the article:

In general, for any specified type of component (with specific physical failure-causing mechanisms), nominally operating at a specific temperature, there is an associated operating temperature increase that will cause the component's failure rate to double. If the temperature increases by 3x this amount, then the failure rate will increase by (approximately) a factor of 2 x 2 x 2 = 8.

If that specific temperature increase is 10C, then the failure rate (i.e. 1/ the expected lifetime) will (approximately) double for every 10C increase in its operating temperature. Likewise, the failure rate will be reduced by (approximately) a factor of 2 for every 10C decrease in its operating temperature.

The temperature increase that causes a component's failure rate to increase by a factor of two is related to a parameter that is specific to the type of component (transformer, semiconductor device, ...) called the "activation energy". The temperature increase (in kelvins or degrees centigrade) that causes a doubling of the failure rate is approximately: 6.24 electron-volts / the activation energy (expressed in electron volts). This assumes that the nominal operating temperature is 50C = 323K.

Example:

If the activation energy associated with a typical transformer is 0.4eV, then the temperature increase required to reduce the expected lifetime of the transformer by a factor a 2 is approximately 6.24eV/0.4eV = 15.6 degrees centigrade.

[Note from AB2EZ: there are a lot of assumptions, built into this theory, about the existence of, and the nature of physical/chemical mechanism(s) that result in an accumulation of microscopic damage, over time... ultimately leading to the failure of a device. For example, the effect of temperature on the rate of growth of a crystalline defect within an electronic device, or the rate of deterioration of the insulation between two adjacent electrical conductors within a transformer, or the formation of a non-conducting film on the surface of an electrical contact. The theory is widely used, and has been shown to be widely useful. However, for some types of failures (e.g. caused by rare, one-time, macroscopic events... like carelessly spilling soup into your laptop), this theory will not be applicable]



Stu
Logged

Stewart ("Stu") Personick. Pictured: (from The New Yorker) "Season's Greetings" looks OK to me. Let's run it by the legal department
W3RSW
Contributing
Member
*
Offline Offline

Posts: 3308


Rick & "Roosevelt"


« Reply #10 on: July 29, 2014, 10:31:16 AM »

May be interesting to note that a lot of component heat failure is because of chemical change and/or loss of volatiles in insulation coatings, impregnated papers, bearing oils and lubes, as well as more thought of stuff like thoriation loss in filaments, loss of vacuum and seals, drying out of electrolyte, metallic contamination  of dielectrics, etc.

Of course, time is one of the essential factors in degradation, the necessary abscissa in the three dimensional display of the big three, degradation vs. time and temperature.  An interesting set of volumetric displays showing linear to exponential changes may be derived from various factors from the design and construction of components, e.g., The "Chinese" versions with rapidly rising failure rates in short time and with little heat change.

Might be fun to see a three dimensional scale standardized for individual or sets of components. -a display of volume pictographs for, say SWL portable receivers.


Logged

RICK  *W3RSW*
W7TFO
WTF-OVER in 7 land Dennis
Contributing
Member
*
Offline Offline

Posts: 2521


IN A TRIODE NO ONE CAN HEAR YOUR SCREEN


WWW
« Reply #11 on: July 29, 2014, 02:13:43 PM »

What I've noticed over the years is transformers built to serve a widely variable load such as pole pigs run hot hence cooling oil & fins or a serpentine.

Transformers designed for a more-or-less fixed load run pretty cool unless there is a downstream problem.

73DG
Logged

Just pacing the Farady cage...
ka4koe
Contributing
Member
*
Offline Offline

Posts: 1157


It's alive. IT'S ALIVE!!!


« Reply #12 on: July 29, 2014, 03:31:47 PM »

Rule of thumb in electrical interior distribution was not to load up to more than 80% nameplate KVA rating (calculated demand load). Take 5% of that and give figure to HVAC guys to calculate heat load. YMMV.
Logged

I'm outta control, plain and simple. Now I have a broadcast transmitter.
pa0ast
Member

Offline Offline

Posts: 62


« Reply #13 on: July 30, 2014, 06:28:11 AM »

From my own expiriance.
If a winding of a transformer conducts some Dc current  than it wil heat up very quickly. If the rectifier tube (s) is working only half,  the current is a half sinus trough the winding and this causes partial saturation of the trafo! then the temp rise can be  huge. The BH curve of the iron shifts to one side and the iron core will go in saturation on one side of the sinus.
Less iron, 60Hz vs 50Hz for instance,  gives more problems in that case. A ringcore transformer is even worse , it can not withstand more than a few miliamp of DC.  I had this problem with my BC610. one of the two rectifiers had a bad filament contact in the tube socket.
RGDS Anton pa0ast
Logged
AB2EZ
Member

Offline Offline

Posts: 1722


"Season's Greetings" looks okay to me...


« Reply #14 on: July 30, 2014, 09:58:19 AM »

Note: after posting this comment... I had some second thoughts about the issue raised here... which are included in my next post in this thread.

Stu


Anton

That's a really important point (in my opinion)!

Saturation of the transformer core, if the rectifier is not drawing the same average current on each half-cycle, is something I never considered.

Having a complete failure of one of the diodes would certainly cause that (and also increased heating of the winding, due to the associated larger peak current on the other half cycle)...

With vacuum tube rectifiers, I think it would be possible for one diode to be delivering significantly less average current to a full wave rectifier, even if both diodes are still working.

Best regards
Stu
Logged

Stewart ("Stu") Personick. Pictured: (from The New Yorker) "Season's Greetings" looks OK to me. Let's run it by the legal department
AB2EZ
Member

Offline Offline

Posts: 1722


"Season's Greetings" looks okay to me...


« Reply #15 on: July 30, 2014, 11:54:24 AM »

Anton
Et al.

The phenomenon you raised is (I think) a lot more complicated than it first appears.

On further thought, I one considers the special case of a simple half wave rectifier:

It is true that the current flowing in the secondary winding will have an average (DC) value. Therefore it will produce an average magnetic field in the core.

But, the current flowing in the primary will also have an average value... and will produce a DC magnetic field in the core that cancels out the DC magnetic field in the core that is produced by the secondary's average current.

Therefore, there will be no net DC magnetic field in the core associated with these two non-zero average currents.

The extra heating of the transformer in a half wave configuration (because one of the two diodes of a full wave rectifier is not functioning) would be associated with the larger amplitude current peaks (for the same average forward current) in a half wave, capacitor input rectifier vs a full wave capacitor input rectifier. The higher current peaks (even though they occur half as often) will produce higher resistive losses in the windings, and higher hysteresis losses in the core (but not due to a DC magnetic field)... because both of these losses increase with the square of the current.

This may be somewhat confusing, because we know that "a transformer cannot pass DC". Perhaps a clearer statement would be that: a transformer cannot produce a non-zero average voltage across any of its windings... because that would imply that the magnetic flux passing through the turns of that winding is increasing toward infinity.

But, a careful analysis would (I believe) show that the "transformers cannot pass DC" constraint is not being violated in the case of an AC voltage source feeding a transformer that feeds a half wave rectifier. The careful analysis would have to be done by taking into account: the time varying magnetic fields in the core produced by the input winding and (separately) by the output winding, the residual (net) magnetic field in the core which produces the time-varying magnetic flux that passes through each winding, and the time-varying amount energy that is stored in the core. This cannot be done using linear system analysis methods (because the load on the secondary of the transformer is non-linear).

Stu
Logged

Stewart ("Stu") Personick. Pictured: (from The New Yorker) "Season's Greetings" looks OK to me. Let's run it by the legal department
W3RSW
Contributing
Member
*
Offline Offline

Posts: 3308


Rick & "Roosevelt"


« Reply #16 on: July 30, 2014, 02:41:36 PM »

Looks to me that in sketching out a full wave circuit with 1/2 wave DC output coupled through a resistance (the load) back to the center tap of the transformer secondary, it is nothing more than a transformer that has one side of secondary active and other side of secondary simply floating (not doing anything) or partially contributing but always 180 degrees opposite phase.  .

DC produced by the rectifier regardless of one half having less emission than the other is DC and will be dissipated by the load before returning to the center tap of the transformer.

No difference between a non-used winding or open secondary than that of the primary which might have several non-used selection taps that are left open.


In a partially open or conducting rectifier half, you can only contribute partially to the working one's half, not some magically enhanced power (heat)  produced beyond what the full transformer rating produces.  Phase will always be 180 degrees opposed in CT windings regardless of the contribution of either half winding by virtue of construction. There is no active component in a transformer that produces amplification (unless of course your looking at magnetic amplifiers, a different beast. )  By 180 deg. opposed superposition your always looking at less that perfect output regardless of each half's contribution.

Decreased supply current across a constant load will simply cause voltage to sag. Such sag will in effect match the delivered current with less heat produced from efficiency losses in all upstream sources and downstream sinks.
The transformer will run cooler simply because less current flows and always as it's supposed to, one winding 180 opposite phase than the other.  How does this generate DC?  Those phases are not isolated, both outputs are being sinked to a device that generates DC and thence to a load that uses that plus any residual AC ripple.

**So don't doubt you Anton, But did the BC 610 transformer actually heat up a lot more? Granted a lot of heat was generated around the Hg Vapor rectifier tube base? 

And Stu, if the phenomenon you describe is so, it would have been noted by just about every service tech, engineer and designer in the last 100 years, and would have been very common and rated right up there as possible/probably hot transformer and transformer failure causes along with failure of electrolytics, etc.   -Particularly noted in early years when tube construction wasn't so standardized and developed, say before the 80, and 5Y3 era.

"Boys, check those dual plate rectifiers for a failed diode along with those nasty electrolytic capacitors if you find a burned out transformer or even one just running hot.  You can't be too careful."
Logged

RICK  *W3RSW*
AB2EZ
Member

Offline Offline

Posts: 1722


"Season's Greetings" looks okay to me...


« Reply #17 on: July 30, 2014, 03:09:39 PM »

Rick

Perhaps we are "talking" about different issues.

If one has a full wave rectifier feeding into a filter capacitor in parallel with a load:

Every half cycle, the load will draw a (roughly) constant current from the capacitor. This will reduce the charge stored in the capacitor, and the output voltage of the supply by some (generally) small percentage.

During some portion of each half cycle, one of the two diodes will conduct a pulse of current that recharges the capacitor.

If one of the two diodes stops working, then the load will still be drawing current in both half cycles (decreasing  the charge on the capacitor and the voltage across the capacitor).

But, in that case, the diode that is still working will have to deliver twice as much charge to recharge the capacitor as it did when both diodes were working.

If you do an LTSpice simulation, you will see that the peak current that will flow through the working diode (once each cycle) will be roughly twice as large as the peak current that flows through each diode (once each cycle) when both diodes are working.

Since the resistive power loss in each half-winding is proportion to i x i, the average power dissipated by the single half-winding in the half wave rectifier will be twice as large as the total average power dissipated by both half-windings in the full wave rectifier.

Therefore there will be twice as much core heating due to resistive losses.

In both cases, half wave rectifier and full wave rectifier, the power delivered to the resistive load will be approximately the same (a little less in the half wave case due to the larger percentage drop in the output voltage between rechargings of the filter capacitor)

In summary: remember that the transformer is delivering pulses of current to recharge the capacitor (once or twice each cycle) even though the load is drawing a nearly constant current.

Stu
Logged

Stewart ("Stu") Personick. Pictured: (from The New Yorker) "Season's Greetings" looks OK to me. Let's run it by the legal department
W3RSW
Contributing
Member
*
Offline Offline

Posts: 3308


Rick & "Roosevelt"


« Reply #18 on: July 30, 2014, 03:28:40 PM »

If one diode stops working then that winding is isolated.  The transformer will deliver only 1/2 the current to anything with the other half wave pulses missing as though it were always a half wave rectifier all along.  This now 1/2 current delivered to an electrolytic in parallel with resistance or resistance only will now be at the same voltage as originally only if the load magically rises in resistance by twice. The electrolytic serves only to smooth out the produced DC, easier in full wave, of course.  All this DC and residual ripple regardless of being produced by half or full wave will be dissipated in the load.

Quote
But, in that case, the diode that is still working will have to deliver twice as much charge to recharge the capacitor as it did when both diodes were working.

The diode still working can't deliver twice the same current and voltage to the load (the AC component of the load affecting the cap)  BECAUSE the other diode's source, one half the transformer's secondary isn't in circuit to provide its share. - Unless that diode is shorted.  Then you'll see heat. The working diode's winding is designed to provide only one half the current; because it can't the voltage will sag hence since Q=CV with V dropping drastically your cap won't be too troubled.

You simply now have a half wave rectifier that due to twice its design load is drawn down in voltage.  The rest of the downstream apparatus will struggle of course.
Logged

RICK  *W3RSW*
AB2EZ
Member

Offline Offline

Posts: 1722


"Season's Greetings" looks okay to me...


« Reply #19 on: July 30, 2014, 03:42:31 PM »

Rick

I guess we will have to agree to disagree on this.

Stu
Logged

Stewart ("Stu") Personick. Pictured: (from The New Yorker) "Season's Greetings" looks OK to me. Let's run it by the legal department
W3RSW
Contributing
Member
*
Offline Offline

Posts: 3308


Rick & "Roosevelt"


« Reply #20 on: July 30, 2014, 04:54:17 PM »

Well l I can understand what your saying and um, upon reflection I see that trying a full load on "half" a winding will produce somewhat more heat for that winding than normal but since voltage will also sag won't be as severe as one would expect. Also the winding not in service since it's rectifier is out or only partially working will be producing no or less heat respectively. The transformer may not even see a net diff. other than local heat. I still think it will see less overall since less power is being drawn total.

I'm not sure what agreeing to disagree means.

Really. I think It's sort of like say "just saying," if you know what I mean.  Grin

We said it; got ourselves and people thinking and hopefully will arrive at some agreement, if even partial and be constructively aided by others.  I tend to overthink things sometimes but try to go to basics.  My formal electronics 'training' is based on physics, thermo, optics and some electronic courses. My informal training is based on some experience over the years.

last dig.-  the cap. will only have to work 1/2 the time at 60 cycles, heh, heh, so can stand a little more charge and discharge.  How about us?

Did you do an LT Spice sim with a real transformer model (wire size, I/R drop, etc.)  or one that simply models winding ratios like the one I use in Circuit Maker, TM ?
Logged

RICK  *W3RSW*
W3RSW
Contributing
Member
*
Offline Offline

Posts: 3308


Rick & "Roosevelt"


« Reply #21 on: July 31, 2014, 09:59:01 AM »

Slept on it and seems I've not connected in thought an increasing load with a dead short as a lower limit.  I stated, " then you'll see heat. " Of course. -Oh, What was I thinking about other halves not doing their share, etc.?
So on the way to that short is an ever lowering resistance load. Nice try Ricardo. Stu, you are right. You get the privilege of saying, "darn, almost wrong again."  Grin

Still, from a service standpoint, almost makes me want to lash up a CT, full wave circuit with only one diode conducting and measure xformer temperatures compared to both diodes working.  Might be fun to see if serious baking occurs, enough to damage a transformer.
Logged

RICK  *W3RSW*
AB2EZ
Member

Offline Offline

Posts: 1722


"Season's Greetings" looks okay to me...


« Reply #22 on: July 31, 2014, 10:24:01 AM »

Rick

Some of this stuff can be very non-intuitive. I must have spent several hours thinking about all of this yesterday.

Yes, it wouldn't be hard to do an experiment by taking a working full wave rectifier, and temporarily disconnecting one end of one of the diodes.

One easy thing to measure will be the impact... of opening up one of the diode branches... on the output voltage of the power supply (under load). The average voltage should drop a few percent, and the peak-to-peak ripple should double.

You could also add a small value resistor (1 ohm) between the secondary center tap and ground. The voltage across that resistor would be 1 ohm x the total current in both halves of the secondary winding. [Note: in either the full wave rectifier or the half wave rectifier, only one half-winding will be conducting current (if any) at any instant of time.]

With both diodes working, you would see (with an oscilloscope) a negative pulse of voltage across the added resistor every half cycle. With only one diode connected, you would see a negative pulse of voltage across the added resistor once every full cycle... but it would have approximately twice the area... and it would have a significantly larger amplitude (as well as a somewhat larger width) v. the full wave case.

If you had a way of measuring the true rms value of the voltage across the added resistor (not just a meter that says it measures true rms values, but that really doesn't) you would see that the true rms value of the voltage across the added resistor (and therefore the true rms current in one or both half-windings... each of which only conducts current when the other is not conducting current) is larger when only one diode is connected. Since only one of the half-windings is conducting current at any instant of time (in either case), the resistance through which this rms current is flowing is always the resistance of one-half winding (not the resistance of two half-windings in parallel). Therefore, if the true rms voltage across the added resistor is larger with the half wave rectifier (as it will be)... there will be more heat produced by resistive losses in the secondary winding (and also in the primary winding).


Best regards
Stu
Logged

Stewart ("Stu") Personick. Pictured: (from The New Yorker) "Season's Greetings" looks OK to me. Let's run it by the legal department
W3RSW
Contributing
Member
*
Offline Offline

Posts: 3308


Rick & "Roosevelt"


« Reply #23 on: July 31, 2014, 06:38:34 PM »

Well I was just going to let the DC load resistor (of sufficient wattage)!tell the tale as much as possible.  E and I, both cases through a 5 or 10 k load resistor. Use a cap of sufficient value, say my big Motorola 500 uf at 450 volts so that any residual AC ripple can be considered in same order as simple error.

I'd use the RMS guessometer figuring relative voltages on the AC side regardless of shape factor to yield data.  I like the idea of a 1 ohm across the CT return for various data.

Any chro/al thermocouple or iron/ Constantine will work, even uncalibrated. Just looking for differences, not absolute values.
Logged

RICK  *W3RSW*
Pages: [1]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.086 seconds with 19 queries.