Author Topic: Trasformer rating for bridge rectifier  (Read 2292 times)

0 Members and 1 Guest are viewing this topic.

Offline taydinTopic starter

  • Frequent Contributor
  • **
  • Posts: 520
  • Country: tr
Trasformer rating for bridge rectifier
« on: January 18, 2019, 02:18:24 pm »
So I wanted to build a 5.2V, 5.25A DC linear power supply, and without doing any detailed calculations, I figured:

1 bridge rectifier, 10A, 600V
1 choke after the bridge rectifiers to help with ripple (didn't have a particular inductance in mind)
2 4700 µF, 25V electrolytic caps in parallel

And for the transformer secondary voltage, I added the output voltage, the diode voltage drops and 2V margin for the linear regulator: 5.2 + 1.4 + 2 = 8.6, or 9V
And for the transformer secondary current, I just arbitrarily chose a current of 7 Amps, just to have a margin.

But then I thought I should do a simulation of this circuit in TINA. Here is what it looks like:



Real programmers use machine code!

My hobby projects http://mekatronik.org/forum
 

Offline taydinTopic starter

  • Frequent Contributor
  • **
  • Posts: 520
  • Country: tr
Re: Trasformer rating for bridge rectifier
« Reply #1 on: January 18, 2019, 02:23:17 pm »
I already had the 4700 µF caps available, so I measured their ESR to be about 10 mΩ. I also had a 5V, 7A transformer lying around, and I measured its winding resistance to be about 24 mΩ. That's the figure I used for my simulated transformer.

The simulated transformer ratio is such that the input is 220Vrms, and output is about 9.2Vrms. I did transient simulation with 0 initial conditions, so the initial high current surge makes sense, and the rectifier should be able to deal with it. But then, during normal operation, the total current coming out of the bridge rectifier, AM3, is peaking at 18 Amps. I measured the average of that waveform, it's about 7.57A, and the RMS of it is 11.74A.

Now I'm not sure how to rate the secondary current of the transformer ... Should I use the average current, or the RMS, or the peak 18A?
« Last Edit: January 18, 2019, 02:29:15 pm by taydin »
Real programmers use machine code!

My hobby projects http://mekatronik.org/forum
 

Offline taydinTopic starter

  • Frequent Contributor
  • **
  • Posts: 520
  • Country: tr
Re: Trasformer rating for bridge rectifier
« Reply #2 on: January 18, 2019, 02:26:53 pm »
If I look at this document:

http://www.hammondmfg.com/pdf/5c007.pdf

It says that the secondary voltage should be 9.55V (it doesn't account for the diode drops and the regulation overhead, I added them), while my simulation says 9.2V is enough.

It says that the secondary current should be 5.25 / 0.94 = 5.58 Amps! But this result doesn't make any sense to me. How can a transformer with 5.58A capability periodically deliver 18A peaks without heating up?
Real programmers use machine code!

My hobby projects http://mekatronik.org/forum
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 20355
  • Country: gb
  • 0999
Re: Trasformer rating for bridge rectifier
« Reply #3 on: January 18, 2019, 02:30:14 pm »
Yes the peaks increase power dissipation. The transformer doesn't need to be rated to 18A continuously, but it certainly needs to be rated to more than 5.25A. Assuming no losses and a huge filter capacitor, the RMS input current to the rectifier, will be 1.414 times the output current, so that's just over 7.4A.

The transformer needs to be a higher voltage, as the voltage will fall, when the filter discharges and the mains voltage could be on the low end of the tolerance range. Go for a 12V 100VA transformer.

Why not just buy a 5V switched mode power supply with an adjustable output voltage? 5V is readily available and getting one with an adjustment potentiometer so the voltage can be set to 5.2V, is no problem. Building a linear regulator is obsolete.
 

Offline taydinTopic starter

  • Frequent Contributor
  • **
  • Posts: 520
  • Country: tr
Re: Trasformer rating for bridge rectifier
« Reply #4 on: January 18, 2019, 02:44:12 pm »
Yes the peaks increase power dissipation. The transformer doesn't need to be rated to 18A continuously, but it certainly needs to be rated to more than 5.25A. Assuming no losses and a huge filter capacitor, the RMS input current to the rectifier, will be 1.414 times the output current, so that's just over 7.4A.

The transformer needs to be a higher voltage, as the voltage will fall, when the filter discharges and the mains voltage could be on the low end of the tolerance range. Go for a 12V 100VA transformer.

Why not just buy a 5V switched mode power supply with an adjustable output voltage? 5V is readily available and getting one with an adjustment potentiometer so the voltage can be set to 5.2V, is no problem. Building a linear regulator is obsolete.

Thanks for the response. A switching power supply certainly makes more sense. And in fact, here is what I'm REALLY trying to do:

I have this HP8592 spectrum analyzer with a dead SMPS power supply. It has the following outputs:

5.2V  5.25A  x1
15V   2.2A    x1
15V   2.0A    x1
12V   1.0A    x1

I tried to fix this power supply, but wasn't successful. The PCB's had extensive electrolyte leakage, tried to clean it up, but it just keeps exploding the main inverter MOSFET's. And with every explosion, the PCB got charred and those locations are also a source of conductance. I want to revisit that and rebuild that power supply from scratch in SMPS topology, but until then, but I'm eager to use the SA, and so I want to build a linear power supply in the meantime.

I tried looking for adjustable SMPS with the necessary capacity but wasn't able to find any locally. But maybe I didn't look hard enough...
Real programmers use machine code!

My hobby projects http://mekatronik.org/forum
 

Offline MrAl

  • Super Contributor
  • ***
  • Posts: 1666
Re: Trasformer rating for bridge rectifier
« Reply #5 on: January 19, 2019, 03:49:48 pm »
Hi,

If you have the waveforms across the device in question you can calculate the average power directly so you dont need to rely on estimates based on RMS or other.  Look up the average power integral definition.
Also, the choke is often better placed on the AC side of the bridge so it does not have to endure the average DC current unless you do that specifically so you can make it swing (which can improve full load efficiency).
 

Offline rstofer

  • Super Contributor
  • ***
  • Posts: 9963
  • Country: us
Re: Trasformer rating for bridge rectifier
« Reply #6 on: January 19, 2019, 04:41:10 pm »
Conduction angle - essentially the amount of time the diode has to conduct all the energy.

As the capacitor voltage gets closer to the maximum rectifier output, the amount of time the diodes are forward conducting gets smaller.  The diodes then have to pass more current in less time and this leads to heating.  From the diode's point of view, ripple is good as it leads to a larger conduction angle (more time to deliver the energy.

Page 16 shows the current waveform as the conduction angle gets small:

ftp://f1006150:robrob99@web.thu.edu.tw/c3.doc

 

Offline taydinTopic starter

  • Frequent Contributor
  • **
  • Posts: 520
  • Country: tr
Re: Trasformer rating for bridge rectifier
« Reply #7 on: January 19, 2019, 05:10:52 pm »
If you have the waveforms across the device in question you can calculate the average power directly so you dont need to rely on estimates based on RMS or other.  Look up the average power integral definition.

Not sure I understand what you meant. I have the current waveform through the load (AM2 in the graph), and can easily have the simulator calculate the average value of it. From there, I can calculate the average power. But that would not take into account the almost 3 times current flowing through the capacitor (see the graph).

I also have the graph of the current coming out of the bridge (AM3 in the graph), and I know the average, RMS and peak values of it. So if I use the average value of AM3 and rate the transformer secondary accordingly, would that transformer be able to deliver 18A for about 8ms, 100 times per second?

Also, the choke is often better placed on the AC side of the bridge so it does not have to endure the average DC current unless you do that specifically so you can make it swing (which can improve full load efficiency).

I wanted to add the choke just to help with ripple, but it came out that the choke limits the rate of increase of the capacitor current. But I might not install it at all, because when I look at the typical "peak forward surge current" specs for 10A bridges, 300A is a common value, so I'm not reall worried about the bridge. I'm only worried about how to rate the transformer.
Real programmers use machine code!

My hobby projects http://mekatronik.org/forum
 

Offline taydinTopic starter

  • Frequent Contributor
  • **
  • Posts: 520
  • Country: tr
Re: Trasformer rating for bridge rectifier
« Reply #8 on: January 19, 2019, 05:19:43 pm »
Conduction angle - essentially the amount of time the diode has to conduct all the energy.

As the capacitor voltage gets closer to the maximum rectifier output, the amount of time the diodes are forward conducting gets smaller.  The diodes then have to pass more current in less time and this leads to heating.  From the diode's point of view, ripple is good as it leads to a larger conduction angle (more time to deliver the energy.

Page 16 shows the current waveform as the conduction angle gets small:

ftp://f1006150:robrob99@web.thu.edu.tw/c3.doc

When the next cycle reaches a voltage that is above the bottom point of the ripple, the capacitor will start charging up. And because the only thing limiting the current is the cap's ESR, transformers's secondary internal resistance and interconnect resistances, that current will be quite high, provided the source can deliver it. While charging the cap, the bridge also has to deliver the load's current. That's why such a high current is observed.

The presence of the choke seems to cause the cap to charge a little slower, but still within 8 ms, which according to my experiments reduces the total current (AM3) from about 21A to 18A.
Real programmers use machine code!

My hobby projects http://mekatronik.org/forum
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 20355
  • Country: gb
  • 0999
Re: Trasformer rating for bridge rectifier
« Reply #9 on: January 19, 2019, 09:17:06 pm »
Yes the peaks increase power dissipation. The transformer doesn't need to be rated to 18A continuously, but it certainly needs to be rated to more than 5.25A. Assuming no losses and a huge filter capacitor, the RMS input current to the rectifier, will be 1.414 times the output current, so that's just over 7.4A.

The transformer needs to be a higher voltage, as the voltage will fall, when the filter discharges and the mains voltage could be on the low end of the tolerance range. Go for a 12V 100VA transformer.

Why not just buy a 5V switched mode power supply with an adjustable output voltage? 5V is readily available and getting one with an adjustment potentiometer so the voltage can be set to 5.2V, is no problem. Building a linear regulator is obsolete.

Thanks for the response. A switching power supply certainly makes more sense. And in fact, here is what I'm REALLY trying to do:

I have this HP8592 spectrum analyzer with a dead SMPS power supply. It has the following outputs:

5.2V  5.25A  x1
15V   2.2A    x1
15V   2.0A    x1
12V   1.0A    x1

I tried to fix this power supply, but wasn't successful. The PCB's had extensive electrolyte leakage, tried to clean it up, but it just keeps exploding the main inverter MOSFET's. And with every explosion, the PCB got charred and those locations are also a source of conductance. I want to revisit that and rebuild that power supply from scratch in SMPS topology, but until then, but I'm eager to use the SA, and so I want to build a linear power supply in the meantime.

I tried looking for adjustable SMPS with the necessary capacity but wasn't able to find any locally. But maybe I didn't look hard enough...
5V at 5.25A should be no problem.
https://www.meanwelldirect.co.uk/wp-content/uploads/sites/55/2018/04/r1748r1748_3.pdf

How about the other voltages? Is that +/-15V?
 

Offline taydinTopic starter

  • Frequent Contributor
  • **
  • Posts: 520
  • Country: tr
Re: Trasformer rating for bridge rectifier
« Reply #10 on: January 19, 2019, 09:26:57 pm »
5V at 5.25A should be no problem.
https://www.meanwelldirect.co.uk/wp-content/uploads/sites/55/2018/04/r1748r1748_3.pdf

How about the other voltages? Is that +/-15V?

Yep. All voltages are with respect to a common ground. Thanks for the link. I'll see if there is a distributor of that company here.
Real programmers use machine code!

My hobby projects http://mekatronik.org/forum
 

Offline taydinTopic starter

  • Frequent Contributor
  • **
  • Posts: 520
  • Country: tr
Re: Trasformer rating for bridge rectifier
« Reply #11 on: January 20, 2019, 10:26:33 am »
5V at 5.25A should be no problem.
https://www.meanwelldirect.co.uk/wp-content/uploads/sites/55/2018/04/r1748r1748_3.pdf

How about the other voltages? Is that +/-15V?

Wow, I looked at this company's product offerings in detail, and there is a power supply that will satisfy almost all the requirements. The model number is QP-150C

https://www.pulsar.pl/pl/karta/QP-150D.pdf

This unit has the following specs:

5V, adjustable, 15A      x1
+15V fixed        3A      x1
-15V  fixed        2A      x1
-5V    fixed     0.6A      x1


I will only use the 5V (adjust it to 5.2V), +15V and -15V. I will obtain the +12V from the +15V using a 7812 fixed regulator. It will still be necessary to make a PCB for this, because the SA requires a line sync signal, which means obtaining a logic level 50 Hz signal that is synchronized to the mains voltage. Also, the power on switch is not standalone. It basically is a logic signal that must turn on power when it is low. I also need a "power good" signal to give to the SA.

Thank you for making me aware of this company. They are very well represented here in Turkey, and the unit costs around 80 Euros.
Real programmers use machine code!

My hobby projects http://mekatronik.org/forum
 

Offline mariush

  • Super Contributor
  • ***
  • Posts: 5170
  • Country: ro
  • .
Re: Trasformer rating for bridge rectifier
« Reply #12 on: January 20, 2019, 12:08:33 pm »
If you want to go all linear, it may make more sense to go with a +/-18v ac transformer , and use switching reg. for 15-25v dc to 5.2v
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf