Author Topic: Transformer question  (Read 4433 times)

0 Members and 1 Guest are viewing this topic.

Offline PinheadBETopic starter

  • Regular Contributor
  • *
  • Posts: 195
  • Country: be
  • Pinball Freak
Transformer question
« on: February 02, 2019, 04:10:08 pm »
Hi,

I just bought a transformer whose datasheet ( https://www.mouser.be/datasheet/2/410/media-1067921.pdf ) states that:
"Primary and secondary windings are designed to be connected in series or parallel. Winding are not
intended to be used independently"

Does that mean that I may not use it to build a +12V/-12V where the currents in each winding would not be perfectly balanced ?
Why is that ?

Thanks
Please keep our planet clean
 

Offline soldar

  • Super Contributor
  • ***
  • Posts: 3595
  • Country: es
Re: Transformer question
« Reply #1 on: February 02, 2019, 04:21:17 pm »
I think that refers to the connection jumpers and I cannot see how it would matter that the windings have equal currents.
All my posts are made with 100% recycled electrons and bare traces of grey matter.
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 6420
  • Country: de
Re: Transformer question
« Reply #2 on: February 02, 2019, 05:31:42 pm »
Very unfortunate phrasing. I think it means you can't use the transformer with two independent 115 V inputs.
Probably added by a lawyer.
It's a perfectly normal transformer.
 

Offline djacobow

  • Super Contributor
  • ***
  • Posts: 1170
  • Country: us
  • takin' it apart since the 70's
Re: Transformer question
« Reply #3 on: February 02, 2019, 05:45:06 pm »
I think they are trying to say that their 4kV isolation guarantee does not exist between the two secondaries, or between the two primaries. So, on the secondary, +12/-12 would be fine; they have the same common, but maybe not ok to generate two totally separate outputs expected to be completely independent.

However, I'd be interested to know how they managed to achieve this limitation!
« Last Edit: February 02, 2019, 06:15:48 pm by djacobow »
 

Offline rcbuck

  • Frequent Contributor
  • **
  • Posts: 355
  • Country: us
Re: Transformer question
« Reply #4 on: February 02, 2019, 05:57:21 pm »
I believe they are trying to say you must use both primary windings. In other words, if you want to use the transformer with a 115 volt input, you cannot use just 1 and 2 or 5 and 6. You probably actually could do that but the output current would only be half of the transformer rating.

You can easily use it for +/-12 volt supply. However, if currents are different for the + and - circuits, the voltage would be slightly more on the circuit that has less current draw. Also, with a light load on the 12 volt windings, you will see voltages closer to 17 volts. If you need +/- voltages to be equal, you will need regulators.

I'm curious what the application is that requires such a large current? Maybe a power amplifier? If so, the current draw should be equal on both + and - sides.
 

Offline djacobow

  • Super Contributor
  • ***
  • Posts: 1170
  • Country: us
  • takin' it apart since the 70's
Re: Transformer question
« Reply #5 on: February 02, 2019, 06:17:51 pm »
I believe they are trying to say you must use both primary windings.
In other words, if you want to use the transformer with a 115 volt input, you cannot use just 1 and 2 or 5 and 6.

This makes the most sense! But man, they could have said that better.
 

Offline Audioguru

  • Super Contributor
  • ***
  • Posts: 1507
  • Country: ca
Re: Transformer question
« Reply #6 on: February 02, 2019, 06:39:02 pm »
The instructions are perfectly clear. Its max rated power is 130VA when it has a 25 degrees C air conditioned room. Less allowed power outdoors in summer.
Primary: Series 230V, Parallel 115V.
Secondary: Series 24V 5.4A, Parallel 12V 10.8A.

You are using the secondaries in series so 24VAC which is 34V peak before rectification and filtering. Then your max allowed current (when the transformer has air conditioning) is 130VA/34V= 3.82A.
The outputs will be unregulated +16V and -16V and about +17V and -17V when not loaded to the max of 3.82A.
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 6420
  • Country: de
Re: Transformer question
« Reply #7 on: February 02, 2019, 07:02:30 pm »
The instructions are perfectly clear. Its max rated power is 130VA when it has a 25 degrees C air conditioned room. Less allowed power outdoors in summer.
Primary: Series 230V, Parallel 115V.
Secondary: Series 24V 5.4A, Parallel 12V 10.8A.

You are using the secondaries in series so 24VAC which is 34V peak before rectification and filtering. Then your max allowed current (when the transformer has air conditioning) is 130VA/34V= 3.82A.
The outputs will be unregulated +16V and -16V and about +17V and -17V when not loaded to the max of 3.82A.

What on earth are you talking about? Mains transformer voltages and currents are always RMS. Your calculations might make sense if the OP wanted to rectify and filter, but this is not stated. And even then they are questionable, as crest factor plays a major role.
 

Offline ArthurDent

  • Super Contributor
  • ***
  • Posts: 1193
  • Country: us
Re: Transformer question
« Reply #8 on: February 02, 2019, 09:34:26 pm »
Benta – “Your calculations might make sense if the OP wanted to rectify and filter, but this is not stated.”
Actually I believe that is what they meant when they said:
PinheadBE – “Does that mean that I may not use it to build a +12V/-12V where the currents in each winding would not be perfectly balanced ?”

AudioGuru and rcbuck had good answers. Here is a handy reference to explain it all.
http://www.hammondmfg.com/pdf/5c007.pdf
 

Offline Brumby

  • Supporter
  • ****
  • Posts: 12413
  • Country: au
Re: Transformer question
« Reply #9 on: February 02, 2019, 10:17:44 pm »
I believe they are trying to say you must use both primary windings.
I would have added: "To get the rated output current"

Quote
In other words, if you want to use the transformer with a 115 volt input, you cannot use just 1 and 2 or 5 and 6.
Yes, just this.  I am lost as to why they would add the secondaries in this statement.  It only makes sense for the primaries.


This makes the most sense! But man, they could have said that better.
Agreed.  They could have simply left the secondaries out of the statement for it to become less confusing - and then added my suggested comment for it to be pretty clear.

Then again, the work that's gone into the engineering of a product is usually more thorough than the documentation.  The only thing I can say is that following the instruction as written will result in proper operation of the transformer.

I think they are trying to say that their 4kV isolation guarantee does not exist between the two secondaries, or between the two primaries. So, on the secondary, +12/-12 would be fine; they have the same common, but maybe not ok to generate two totally separate outputs expected to be completely independent.
While I tend to agree with your conclusion, I'm not too sure they are being explicit enough in the datasheet.  In the statement on Safety, they say: "These units are designed with 4000VAC isolation between the primary and secondary, and also, between each winding and the core." which leaves as much unsaid as said.  In some applications, knowing the rated isolation between the two secondaries would be useful.  The isolation between the two primaries - not so much.


While this datasheet will provide sufficient information for typical applications, it's a bit lacking as far as providing full engineering specifications - IMHO.
« Last Edit: February 02, 2019, 10:20:28 pm by Brumby »
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 6420
  • Country: de
Re: Transformer question
« Reply #10 on: February 02, 2019, 10:31:19 pm »
Here is a handy reference to explain it all.
http://www.hammondmfg.com/pdf/5c007.pdf

That "handy reference" doesn't explain anything at all. Quite the opposite
Please explain to me why "FULL WAVE BRIDGE Resistive Load" tells me to multiply voltage and current by 0.90?
I don't even want to comment on the other configurations, that would be a total waste of time.

Apparently "Hammond Manufacturing" needs any excuse to derate their transformers, leading me to doubt their products.
 

Offline Nusa

  • Super Contributor
  • ***
  • Posts: 2418
  • Country: us
Re: Transformer question
« Reply #11 on: February 02, 2019, 10:32:00 pm »
Quote
3. Secondary: Series: 24.0V CT@ 5.4A; Parallel: 12.0V @ 10.8A

The CT presumably stands for center-tap, so clearly they're perfectly fine with that configuration. That's what you want anyway, right?
 

Offline soldar

  • Super Contributor
  • ***
  • Posts: 3595
  • Country: es
Re: Transformer question
« Reply #12 on: February 02, 2019, 11:19:59 pm »
That "handy reference" doesn't explain anything at all. Quite the opposite
Please explain to me why "FULL WAVE BRIDGE Resistive Load" tells me to multiply voltage and current by 0.90?

VA or Amps are given but they are for a power factor of 1. If power factor is less then Amps should be reduced accordingly. Full bridge rectifiers have a very poor power factor. Thus the 0.90 reduction which I would probably reduce even more.
All my posts are made with 100% recycled electrons and bare traces of grey matter.
 

Offline ArthurDent

  • Super Contributor
  • ***
  • Posts: 1193
  • Country: us
Re: Transformer question
« Reply #13 on: February 03, 2019, 02:36:13 am »
Benta – “That "handy reference" doesn't explain anything at all. Quite the opposite
Please explain to me why "FULL WAVE BRIDGE Resistive Load" tells me to multiply voltage and current by 0.90?
I don't even want to comment on the other configurations, that would be a total waste of time.”

So far you’ve questioned Triad and Hammond (who have been making transformers for ages) about their transformer application knowledge. Perhaps reading other explanations about reducing transformer ratings for use with different rectifier setups might help you understand. The ratios may vary depending on the transformer design used but the principles are the same. 
https://www.powervolt.com/how-to-determine-page-25
https://www.nutsvolts.com/questions-and-answers/transformer-ratings
 

Offline basinstreetdesign

  • Frequent Contributor
  • **
  • Posts: 466
  • Country: ca
Re: Transformer question
« Reply #14 on: February 03, 2019, 06:51:24 am »
Hi,

I just bought a transformer whose datasheet ( https://www.mouser.be/datasheet/2/410/media-1067921.pdf ) states that:
"Primary and secondary windings are designed to be connected in series or parallel. Winding are not
intended to be used independently"

Does that mean that I may not use it to build a +12V/-12V where the currents in each winding would not be perfectly balanced ?
Why is that ?

Thanks

Wow.  Here I thought it was pretty obvious what the manufacturer was referring to.  I never thought a bunch of people like you guys could go off on so many tangents.

The only explanation for that quote that makes sense to me is that to obtain the rated maximum VA rating of the transformer one must use all of the windings.  In parallel or series as needed.  That trying to get that much power through the transformer by using only two windings and leaving the other unconnected would not work so well as the wire gauge would be inadequate for it.  Yes, I'm making an assumption here but it is the only way this makes sense.
STAND BACK!  I'm going to try SCIENCE!
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 3034
  • Country: au
Re: Transformer question
« Reply #15 on: February 03, 2019, 07:16:44 am »
Here is a handy reference to explain it all.
http://www.hammondmfg.com/pdf/5c007.pdf

That "handy reference" doesn't explain anything at all. Quite the opposite
Please explain to me why "FULL WAVE BRIDGE Resistive Load" tells me to multiply voltage and current by 0.90?
I don't even want to comment on the other configurations, that would be a total waste of time.

Apparently "Hammond Manufacturing" needs any excuse to derate their transformers, leading me to doubt their products.
The "0.90" is used to multiply by the RMS voltage to get the Average voltage, 0.9 x 0.707 = 0.636.
I have some doubts about some of the information in the Hammond guide.
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline Ian.M

  • Super Contributor
  • ***
  • Posts: 13216
Re: Transformer question
« Reply #16 on: February 03, 2019, 07:55:12 am »
I have some doubts about some of the information in the Hammond guide.
Which specific circuit(s)?  Every time I've set up a LTspice sim for one of them, the worst case derating factors come out very close to the Hammond figures.  I must admit I haven't tried them all though.
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 3034
  • Country: au
Re: Transformer question
« Reply #17 on: February 03, 2019, 08:38:07 am »
For the FULL WAVE BRIDGE Capacitor Input Load.
V (Avg) D.C. = 0.90 X Sec. V A.C.
With a 30VAC secondary, I wouldn't expect the average DC to be loaded back to 27V at the recommended load derating factor. That's a long way down from the unloaded DC of 42V.

I have started a linear bench supply project so I guess I'll have a better idea when it's fully tested.

For the FULL WAVE BRIDGE Choke Input Load.
"V (Avg) D.C. = 0.90 X Sec. V A.C" mathematically agrees with the expected DC voltage at some light load that causes continuous inductor current flow, assuming that the inductor is large enough.
At full rated load, the DC voltage would be something less due to expected resistive losses.
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline Ian.M

  • Super Contributor
  • ***
  • Posts: 13216
Re: Transformer question
« Reply #18 on: February 03, 2019, 09:24:06 am »
I hadn't noticed that.  |O   A somewhat simplistic LTspice sim of the FULL WAVE BRIDGE Capacitor Input Load circuit that assumes all the transformer losses are resistive is attached.  Its got a lot of .measure commands so view the spice error log for results.

With ideal diodes, a transformer with 10% regulation and 1000uF per A of load current, I get  38.6V average, and 35.8V ripple trough.

It also  shows the Hammond 0.62 current derating factor may be insufficient for 24/7 duty if the transformer has good regulation, and the diodes and reservoir cap are large for the load current.
 

Offline soldar

  • Super Contributor
  • ***
  • Posts: 3595
  • Country: es
Re: Transformer question
« Reply #19 on: February 03, 2019, 09:54:35 am »
If you put the mains voltage on the screen of your scope the first thing you will notice is that it is not a sine at all but the tops are flattened and that is due to all the rectifiers getting shots of current only at the top of the wave.  The output resistance of the network makes the voltage fall.

If you put the current on the scope you will see short bursts of current and zero current most of the cycle.  Power factor is very poor.

The behavior depends an many factors. Higher resistance and lower capacity mean a longer diode conductive duty cycle, a better power factor but worse regulation. OTOH, big capacitor and low impedance means more stable output regulation, lower ripple, but worse power factor, bigger transformer and diodes needed, etc.

Calculations are good but there is a lot of rule of thumb and pure educated estimation. If you are building a single unit the best way is to start building and testing behavior under different loads. 

But, yeah, DC current out of the rectifier + capacitor is nowhere near what the transformer can supply with unity power factor on, say, a purely resistive load.
All my posts are made with 100% recycled electrons and bare traces of grey matter.
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 3034
  • Country: au
Re: Transformer question
« Reply #20 on: February 03, 2019, 07:31:33 pm »
The "V (Avg) D.C. = 0.90 X Sec. V A.C." isn't really a mistake.
It is stating that full wave rectified AC voltage averages 0.9 x RMS voltage.
It isn't referring to the capacitor voltage at all, so it's just irrelevant to the  FULL WAVE BRIDGE Capacitor Input Load. And misleading.
 
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 6420
  • Country: de
Re: Transformer question
« Reply #21 on: February 03, 2019, 07:45:44 pm »
So far you’ve questioned Triad and Hammond (who have been making transformers for ages) about their transformer application knowledge. Perhaps reading other explanations about reducing transformer ratings for use with different rectifier setups might help you understand. The ratios may vary depending on the transformer design used but the principles are the same. 
https://www.powervolt.com/how-to-determine-page-25
https://www.nutsvolts.com/questions-and-answers/transformer-ratings

Apparently you completely missed the "Resistive Load" part. The transformer is specified for resistive load (for complex loads you need reduction factors, of course).
Please explain to me why voltage and current need to be multiplied by 0.9 when inserting a rectifier bridge between transformer and resistive load.

 

Offline soldar

  • Super Contributor
  • ***
  • Posts: 3595
  • Country: es
Re: Transformer question
« Reply #22 on: February 03, 2019, 07:51:48 pm »
Maybe I did not explain it well. Yes, we have all understood by now that in a sine signal RMS value is 0.707 of peak and average value is .637 of peak and therefore average is 0.9 of RMS. It has been explained, understood and accepted by all (I think).

My point is that, further to that, most uses are for rectified and capacitor-filtered power and many people do not understand that a transformer specified as 10 A secondary, that means 10 A RMS, 9 A average, into a pure resistive load but much less than that through a rectifier bridge and capacitor. Many people seem to think that if they need 10 A dc out of the unregulated power supply all they need to order is a "10 A" transformer.
All my posts are made with 100% recycled electrons and bare traces of grey matter.
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 6420
  • Country: de
Re: Transformer question
« Reply #23 on: February 03, 2019, 08:28:32 pm »
Soldar, we understand perfectly.
But really, no one uses average for anything at all, it's a useless measure.
If you want to calculate transformer power loss, RMS is the variable, and yes, with capacitive storage after the rectifier, you need to take crest factor into account, which is considerable when calculating power dissipation in the transformer.

That still doesn't explain why Hammond uses a 0.90 factor for a resistive load.
I'm still waiting for an explanation from Hammond or ArthurDent.

 

Offline soldar

  • Super Contributor
  • ***
  • Posts: 3595
  • Country: es
Re: Transformer question
« Reply #24 on: February 03, 2019, 08:49:27 pm »
Well, I can't say as I have not seen the original or the context. It could just be a plain mistake and I wouldn't give it much importance. Mistakes creep in in any work with more than a few pages. It is inevitable. I go over anything I write several times and later I still find mistakes. It happens in the best regulated families.

Some fifteen years ago I downloaded a document from Texas Instruments and it had so many mistakes I felt compelled to email the author who quickly replied with a tirade telling me I got it for free and had no right to complain. I emailed TI and they quickly replied with an apology and I guess they told the author of the document to STFU because that was the end of it.
All my posts are made with 100% recycled electrons and bare traces of grey matter.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf