Author Topic: Transformer specs. (Answered)  (Read 5920 times)

0 Members and 1 Guest are viewing this topic.

Offline MUDGELTopic starter

  • Contributor
  • Posts: 14
  • Country: au
Transformer specs. (Answered)
« on: November 16, 2018, 04:38:06 pm »
I want to build a linear power supply.

I’m in Australia so mains voltage is 240 VAC, 10 Amps. I want to have an output of 30 VDC, 5 Amps.

What transformer specs are required to meet my desired output? I did a google search using a variety of the keywords used here but don’t understand about Volt Amps and their significance in spec’ing my desired transformer.

By all means just tell me to get “xyz” transformer or teach me how to work it out for myself.

I hope I asked the correct way. 

Thanks. Mike V.
« Last Edit: November 17, 2018, 08:52:03 am by MUDGEL »
I'm retired but still actively running a project recording studio.
Hobby as per above and now including tinkering with electronics.
 

Offline Jwillis

  • Super Contributor
  • ***
  • Posts: 1725
  • Country: ca
Re: Transformer specs.
« Reply #1 on: November 16, 2018, 05:49:09 pm »
 Single phase transformers are wired according to supply either 120 Vac (2 primary connections with single winding),240,Vac (2 primary connection with single winding)or both (4 primary connections with 2 windings). The 4 connection transformers are wired either in parallel for 120Vac or in series for 240Vac.The resistance of both winding's should be equal or very close. The output secondary will have 2 or more connections depending on how its wound and how many individual winding's there are  .Since you only need 30 Vdc at 5 amps you need a transformer with 2 output connections (single winding )with 150 VA minimum and usually is marked according to output(secondary ratings)

S(VA) = I(A) × V(V)

S is apparent power measured in  Volt Amps (VA). I(A) is required out put Amps and (V) is required output Volts.

Generally you want to have a bit more Volts amps to compensate for losses so I would suggest 200 VA -300 VA at 30Volts.I insures you don't over load the transformer at full load.  Having a higher VA won't effect how you power supply operates.But it will be more stable.

Hope this helps.
 

Offline fourtytwo42

  • Super Contributor
  • ***
  • Posts: 1217
  • Country: gb
  • Interested in all things green/ECO NOT political
Re: Transformer specs.
« Reply #2 on: November 16, 2018, 06:31:49 pm »
I want to build a linear power supply.
I’m in Australia so mains voltage is 240 VAC, 10 Amps. I want to have an output of 30 VDC, 5 Amps.

The key issue here is you want a DC output, so you will have to rectify and filter the AC output of the transformer depending upon the requirements of your load. So in order to help you fully we need some information about the load.

The rectifier and filter will also effect the required transformer secondary voltage to achieve 30V DC output so is rather important to your transformer specification.

I assume were talking about an unregulated supply here so the operating voltage range of the load also needs to be taken into consideration.
 

Offline Audioguru

  • Super Contributor
  • ***
  • Posts: 1507
  • Country: ca
Re: Transformer specs.
« Reply #3 on: November 16, 2018, 06:47:15 pm »
We need exact specifications, not assumptions.
30V at 5A is a high power of 150W. What will this power supply power?
1) Do you need the voltage to be variable with a voltage adjustment?
2) Do you need the voltage to be regulated?
3) Do you need the maximum current to be regulated with variable adjustment or simple 5A fixed current limiting be OK?
 

Offline paulflan79

  • Newbie
  • Posts: 8
  • Country: gb
Re: Transformer specs.
« Reply #4 on: November 16, 2018, 06:59:10 pm »
As stated above, more information is needed as the transformer will provide an AC voltage output that will then need to be rectified and smoothing capacitor's used to make a stable DC voltage however this DC voltage will be higher than the AC voltage coming from the transformer depending on the load and and it's demand's on the power supple and the size of filtering capacitor's used however normally speaking, the higher the better if you've already got them ready to use,

here's a common pic of a full bridge rectifier you will need to use with the transformer,

https://www.dropbox.com/s/6hoiulykbf78iwx/full%20bridge%20rectifier%201.jpg?dl=0

however a quick Google search will provide more information on this matter

Here's a starting point...

https://www.google.com.au/search?source=hp&ei=PhPvW7P8I8jDwQKxiq-wBA&q=increased+VDC+when+rectifying+a+transformer&oq=increased+VDC+when+rectifying+a+transformer&gs_l=psy-ab.12...3864.3864.0.5013.1.1.0.0.0.0.143.143.0j1.1.0....0...1c.2.64.psy-ab..0.0.0....0.CH-_8iaTSxQ
 

Offline MUDGELTopic starter

  • Contributor
  • Posts: 14
  • Country: au
Re: Transformer specs.
« Reply #5 on: November 16, 2018, 10:26:14 pm »
Sorry! I neglected to mention that the transformer will be used in a lab bench power supply.

The range is 0-30VDC in 0.01 Volt inc/decrement & 0-5A in 0.01A inc/decrement. If I can do it I would like to be able to provide miilivolt, milliamp values at least below 1V, 1A if not the full range.

I’m not asking for instructions on a whole build just the transformer specs or how to derive them.

Thanks for all the useful input thus far. 

Mike V
I'm retired but still actively running a project recording studio.
Hobby as per above and now including tinkering with electronics.
 

Offline rstofer

  • Super Contributor
  • ***
  • Posts: 9963
  • Country: us
Re: Transformer specs.
« Reply #6 on: November 16, 2018, 11:43:55 pm »
Sorry! I neglected to mention that the transformer will be used in a lab bench power supply.

The range is 0-30VDC in 0.01 Volt inc/decrement & 0-5A in 0.01A inc/decrement. If I can do it I would like to be able to provide miilivolt, milliamp values at least below 1V, 1A if not the full range.

I’m not asking for instructions on a whole build just the transformer specs or how to derive them.

Thanks for all the useful input thus far. 

Mike V

You're talking about a pretty serious PS - especially now that the current has been kicked up to 5A.
You want to make 30VDC - that means you need to drop something across the regulator, say 6V so the DC voltage input needs to be 36V.  As a first approximation, the DC output of a full wave rectifier is 1.4 time the RMS AC voltage so you need a transformer in the range of 26V.

You're talking about 5A at 26V or 130 Watts.  The VA rating needs to account for the power factor of the rectifier and that number probably isn't known.  So, maybe 200 VA would do it.

That should get you in the ballpark.

HOWEVER...  When you try to output 1V at 5A, the regulator has to drop (36V - 1V) * 5A or around 175 Watts - quite a space heater.

And this is why commercial power supplies have multi-tapped transformers with the appropriate tap chosen by relays based on output voltage setting.

I can't seem to build a direct link but you can Google 'mouser 530-MPI-200-28' and get to a 28V 200 VA transformer but the max primary is 230V so the secondary will be a little higher if you drive it with 240V.  You can just go to mouser.com and enter the part number if the search doesn't work.

This power supply topic comes up every week.  The problem is, the specs are not realistic.  5A is pretty high for a general purpose supply.  Why 30V?  Just because Rigol can do it?  Even my Rigol DP832 can't output 5A at 30V.  It's a $500 supply with dual 30V/3A and a single 5V/3A capability.  But it's $500!  And I will never use the higher currents at 30V and, since I don't play with 74xx logic, I probably won't get to 3A on 5V either.  The other day I brought up a complete Z80 system including the CP/M operating system and I think the system took less than 200 mA.  It uses 74LSxx logic plus a CMOS processor chip.

Transformer specs don't favor the DIYer.  It's easy to get a multi-tap toroidal transformer if you want to design it and buy 1000 of them.

I bought the DP832 because I could, not because I needed it.  I can get 5V/2A wall warts all day long.  I can get 12V wall warts or 6V.  If I wanted to, I could build a +-15V supply for op amp projects with 7815 and 7915 1A regulators.

Here is our host's video series re: a lab power supply:


Here is another of our host's power supply projects:
https://www.eevblog.com/projects/usupply/

These two series may give you some ideas.
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 3034
  • Country: au
Re: Transformer specs.
« Reply #7 on: November 17, 2018, 01:05:58 am »
Usually this guide gets referenced , http://www.hammondmfg.com/pdf/5c007.pdf
For a bridge rectifier and capacitative load, the current derating is 0.62. This would be due to the poor power factor caused by loading of the voltage peaks.
I'm not certain what secondary voltage should be chosen. 36Vac rectified and filtered is about 50Vdc unloaded. But what will  it be at 5A load?
Btw, element14 have a good range of transformers.
My Agilent U8002A 30V, 5A power supply from memory has a 40Vac, 10A transfomer, but it also has a phase controlled pre-regulator so  the series pass MOSFET has only 5V across it regardless of the voltage output setting.

« Last Edit: November 17, 2018, 01:07:37 am by xavier60 »
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 3034
  • Country: au
Re: Transformer specs.
« Reply #8 on: November 17, 2018, 01:26:49 am »
I call a choice between these two.
VTX-146-300-215
VTX-146-300-218
Because the 2x15V model is rated at 10A, maybe it will hold up. Can someone simulate it?
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline ArthurDent

  • Super Contributor
  • ***
  • Posts: 1193
  • Country: us
Re: Transformer specs.
« Reply #9 on: November 17, 2018, 02:39:08 am »
Mudgel - "The range is 0-30VDC in 0.01 Volt inc/decrement & 0-5A in 0.01A inc/decrement."

Most commercially sold power supplies can't meet these specs. To get a new one that does meet these specs could cost you up to 4-digits so my advice is to be realistic and forget this project. You don't seem to have the basic theory mastered or know the terminology so a power supply like this is way beyond what you should attempt.  I wouldn't want to do it and actually have bought a supply with those kind of specs for my own use.
 

Offline spec

  • Frequent Contributor
  • **
  • Posts: 833
  • Country: england
  • MALE
Re: Transformer specs.
« Reply #10 on: November 17, 2018, 07:11:28 am »
I want to build a linear power supply.

I’m in Australia so mains voltage is 240 VAC, 10 Amps. I want to have an output of 30 VDC, 5 Amps.

What transformer specs are required to meet my desired output? I did a google search using a variety of the keywords used here but don’t understand about Volt Amps and their significance in spec’ing my desired transformer.

By all means just tell me to get “xyz” transformer or teach me how to work it out for myself.

I hope I asked the correct way. 

Thanks. Mike V.

Hi Mike,

That is some lab PSU you are aiming at 8)

Can I ask why you want to build it. Is it just for the experience or do you just need a power supply to use?

If it is for the latter, as the other members have indicated, it is much better to buy one. You can get a 0V to 30V, 0V to 10A accurate linear lab PSU with digital read out for £50UK or the equivalent in Oz (a guy on another forum from NZ bought one and I have three). 

To answer your question about the transformer specification here it is: 40V RMS at 5A RMS if you go for the common E/H transformer. That means a 200W rating. But I would advise to get two 20V secondaries at 5A (see pre switching below).
It does not matter if the two secondary windings are connected or not.

But, best go for a toroidal transformer. They have a better electrical performance, are more compact, and have less stray magnetic fields. With a toroidal transformer you may have to limit the inrush current on the primary at switch on, but that is a simple matter.

I suppose you know that you will need a pretty hefty reservoir capacitor to keep the ripple voltage down and to handle the ripple current.

You will also need a very low thermal resistance heatsink with additional fan cooling.

Without any kind of primary supply switching you will need six output power transistors. With 50% pre switching you will need three output power transistors.

A good  bridge rectifier spec is : 20A upwards at 100V upwards. If you are wondering why 20A for a 5A supply, it is to limit the forward voltage drop when the huge gulps of current are charging the reservoir capacitor. If you can make a bridge rectifier from Schottky diodes, even better.

If you want to know how to calculate/guesstimate the transformer parameters, the reservoir capacitor parameters, and even the transistor/heatsink parameters just shout and I will run through them.  It is pretty boring stuff though. :palm:
« Last Edit: November 17, 2018, 08:01:56 am by spec »
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 12537
  • Country: us
Re: Transformer specs.
« Reply #11 on: November 17, 2018, 07:36:56 am »
Sorry! I neglected to mention that the transformer will be used in a lab bench power supply.

The range is 0-30VDC in 0.01 Volt inc/decrement & 0-5A in 0.01A inc/decrement. If I can do it I would like to be able to provide miilivolt, milliamp values at least below 1V, 1A if not the full range.

I’m not asking for instructions on a whole build just the transformer specs or how to derive them.

Thanks for all the useful input thus far. 

Mike V

As someone already mentioned, a lab power supply with these specs purchased off the shelf would cost you an arm and a leg, and maybe a kidney too. It would have taken expert electronics engineers many hours to design and test and debug.

This project is way off the scale for a hobby build. You do not seem to understand what you are contemplating.

If you want to succeed I recommend you to set your sights a bit lower. Go for something like 0-15 V and 0-1 A and you will have a much better chance of success.
 

Offline MUDGELTopic starter

  • Contributor
  • Posts: 14
  • Country: au
Re: Transformer specs.
« Reply #12 on: November 17, 2018, 08:51:11 am »
I already have a cheap and cheerful Chinese made 0-30V, 0-5A linear power supply. Korad KA3005P

It works fine,  especially for the price.

I want to challenge myself not by assembling a kit but actually building an LPS from the ground up.

Thanks to those who helped and encouraged me. I think I have what I need now. Clearly I have a lot to learn so I’ll go continue doing that now.

I'm retired but still actively running a project recording studio.
Hobby as per above and now including tinkering with electronics.
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 12537
  • Country: us
Re: Transformer specs. (Answered)
« Reply #13 on: November 17, 2018, 03:19:14 pm »
I already have a cheap and cheerful Chinese made 0-30V, 0-5A linear power supply. Korad KA3005P

It works fine,  especially for the price.

I want to challenge myself not by assembling a kit but actually building an LPS from the ground up.

Thanks to those who helped and encouraged me. I think I have what I need now. Clearly I have a lot to learn so I’ll go continue doing that now.

OK, cool. You could learn a lot about how to build such a thing by disassembling your Korad and reverse engineering it. Look at the transformer and see what windings and voltages it has, trace out the circuit board and draw the schematic, look at what kind of pass elements are used, what size heat sinks, what cooling fans and air flow arrangements, etc.

Electrically, look at the precision at which you can set and regulate the voltage and current. If you have an oscilloscope look at the regulation under different kinds of loads and load changes, also the regulation and response time at switch on and switch off. See if you can find what limitations it has that let it be designed down to the price.
 

Offline mariush

  • Super Contributor
  • ***
  • Posts: 5170
  • Country: ro
  • .
Re: Transformer specs. (Answered)
« Reply #14 on: November 17, 2018, 03:35:21 pm »
if you want to make something fancy, you can replace the bridge rectifier with a LT4320 and four mosfets : https://www.analog.com/media/en/technical-documentation/data-sheets/4320fb.pdf

Saves you around 8-10 watts of heat when you're close to 5A of output, and around 1-2v of voltage drop (that would be across the diodes in the rectifier).

As for the transformer, yeah, see the post above with the hammond link ... use 0.62 for a reasonable approximation.

If you want 30v out and 5A, then you're looking at around 35v dc before going through transistors so let's say a peak of 40v dc ... that's around 28v AC ... so you'll need a 2 x 15v AC transformer or a 2x18v AC transformer.  So you want 5A ... with the 0.62 factor, that means you need something around 8A .... so you're looking at 2x15v 8A = ~ 240VA (round it up to 250VA)  or 2x18v 8A  = 288VA (round it up to 300VA)

So either of those two you mentioned in a previous post would work, if they're 300VA.

You'll probably also want a separate 2 x 15-18v ac transformer for +/-15v for opamps, 12v, 5v etc
 
The following users thanked this post: spec

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 3034
  • Country: au
Re: Transformer specs. (Answered)
« Reply #15 on: November 17, 2018, 11:07:28 pm »
The elimination of diode voltage drop with the use of the LT4320 and four mosfets will make the 2 x 15V transformer more likely to maintain regulation at full output voltage and current.
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline MUDGELTopic starter

  • Contributor
  • Posts: 14
  • Country: au
Re: Transformer specs. (Answered)
« Reply #16 on: November 18, 2018, 05:31:31 am »
Thanks very much for your help everyone. To those for technical help and others for helping a newbie set more realistic goals.
I'm retired but still actively running a project recording studio.
Hobby as per above and now including tinkering with electronics.
 

Offline spec

  • Frequent Contributor
  • **
  • Posts: 833
  • Country: england
  • MALE
Re: Transformer specs. (Answered)
« Reply #17 on: November 18, 2018, 06:10:06 am »
if you want to make something fancy, you can replace the bridge rectifier with a LT4320 and four mosfets : https://www.analog.com/media/en/technical-documentation/data-sheets/4320fb.pdf

Saves you around 8-10 watts of heat when you're close to 5A of output, and around 1-2v of voltage drop (that would be across the diodes in the rectifier).

Hi Mariush,

That MOSFET driver chip is pretty neat- thanks for posting :-+ Have you ever used one. If so, are they well behaved?
« Last Edit: November 18, 2018, 06:15:52 am by spec »
 

Offline spec

  • Frequent Contributor
  • **
  • Posts: 833
  • Country: england
  • MALE
Re: Transformer specs. (Answered)
« Reply #18 on: November 18, 2018, 06:13:47 am »
Thanks very much for your help everyone. To those for technical help and others for helping a newbie set more realistic goals.

No sweat from me :)  and good luck with your power supply, whichever approach you decide on
 

Offline mariush

  • Super Contributor
  • ***
  • Posts: 5170
  • Country: ro
  • .
Re: Transformer specs. (Answered)
« Reply #19 on: November 18, 2018, 07:16:50 pm »
if you want to make something fancy, you can replace the bridge rectifier with a LT4320 and four mosfets : https://www.analog.com/media/en/technical-documentation/data-sheets/4320fb.pdf

Saves you around 8-10 watts of heat when you're close to 5A of output, and around 1-2v of voltage drop (that would be across the diodes in the rectifier).

Hi Mariush,

That MOSFET driver chip is pretty neat- thanks for posting :-+ Have you ever used one. If so, are they well behaved?

I didn't test one PERSONALLY, but several members on this forum have tested it and had only good things to say about it (at least from what I remember reading on forum). You'll also find plenty of other reviews on various forums about it, from memory pretty much all comments were positive (minus the comments about the price)

It is a Linear part (they were bought by Analog Devices) and Linear parts were even before more expensive compared to parts from other manufacturers, but they're also good parts with good to very good documentation through datasheets and/or application notes.

I suspect the part isn't more popular because of its limitation: it supports maximum 72v AC ... so it kinda sucks that you can't use it on mains 110v AC or 230v AC but for your needs, to convert 30-40v AC to DC, it would be perfectly suitable.

A lot of people used these to replace bridge rectifiers in older class AB audio amplifiers - those extra 1.5v-2v can make a difference there if the transformer used or the bulk capacitors are undersized. Can add a few watts of audio power in the end.
 

Offline spec

  • Frequent Contributor
  • **
  • Posts: 833
  • Country: england
  • MALE
Re: Transformer specs. (Answered)
« Reply #20 on: November 18, 2018, 07:36:13 pm »
Thanks for the info- very useful :)
 

Offline not1xor1

  • Frequent Contributor
  • **
  • Posts: 716
  • Country: it
Re: Transformer specs. (Answered)
« Reply #21 on: November 19, 2018, 07:27:59 am »
The elimination of diode voltage drop with the use of the LT4320 and four mosfets will make the 2 x 15V transformer more likely to maintain regulation at full output voltage and current.

another member of this forum, Liv, designed some very good PSUs using just a couple of LM393 as driver for the mosfet rectifier bridge.
https://www.eevblog.com/forum/projects/diy-bench-power-supply-psl-3604/msg623000/#msg623000
LT4320s are relatively expensive.
 
The following users thanked this post: spec

Offline mariush

  • Super Contributor
  • ***
  • Posts: 5170
  • Country: ro
  • .
Re: Transformer specs. (Answered)
« Reply #22 on: November 19, 2018, 08:33:39 pm »
They're not THAT expensive.

LT4320 is around 6$ if you buy 1, 4$ if you buy 25
A good bridge rectifier (rated for 10-20A) is around 1$ if you buy 1, a heatsink for it would be another 0.5$ ... so you're looking at 1.5$ vs 6$

Those 20-30 components he uses to build the equivalent of the LT4320 would probably cost 2.5-3$ if bought from a distributor, simply because you're ordering small quantities of a lot of different parts. He already had some of the parts before, maybe he ordered some components locally... whatever.
Then there's also the amount of space used on the circuit board... but circuit boards are cheaper these days.. the extra circuit board space is probably less than 1$, depending on where you order the board from.   

That guy's problem was that he's living in Belarus, and therefore ordering online is difficult and shipping to him is costly. For these reasons, he had to work with parts he can easily purchase locally or pull from other hardware.

For others, shipping is free  or cheap... for example farnell and rs-components and tme.eu have very reasonable shipping costs to MY country (Romania), something like 4-5$ per order, regardless of how big the package is. So, if someone orders all the parts for that power supply, or orders 100-200$ worth of electronic components, paying less than 5$ on shipping is not a big deal. 

Even Digikey has free shipping from US if you order 60$ or more (well, it was like this around 1 year ago when I last ordered) but I admit it was an annoyance having to go to the local DHL office (at the airport, at the outskirts of my town, around 5 miles/ 8 km) to pay around 2-3% of package value in customs fees and 20% in VAT.  Maybe things have changed since then but these days I mostly order from farnell and tme.eu
 


 
The following users thanked this post: spec

Offline not1xor1

  • Frequent Contributor
  • **
  • Posts: 716
  • Country: it
Re: Transformer specs. (Answered)
« Reply #23 on: November 20, 2018, 07:59:51 am »
They're not THAT expensive.

LT4320 is around 6$ if you buy 1, 4$ if you buy 25
A good bridge rectifier (rated for 10-20A) is around 1$ if you buy 1, a heatsink for it would be another 0.5$ ... so you're looking at 1.5$ vs 6$

Those 20-30 components he uses to build the equivalent of the LT4320 would probably cost 2.5-3$ if bought from a distributor, simply because you're ordering small quantities of a lot of different parts. He already had some of the parts before, maybe he ordered some components locally... whatever.
Then there's also the amount of space used on the circuit board... but circuit boards are cheaper these days.. the extra circuit board space is probably less than 1$, depending on where you order the board from.
[...]

well, we are dealing with facts not opinions, so let's do some fact checking...

LT4320 vs. bridge
tme Italian price (VAT excl.):
LT4320                4.73€  @ 10 pieces
400V-20A bridge   0.58€  @ 5 pieces (0.50€ @ 25 pieces)

Besides that you have to spend at least a couple of euros to get the power mosfets, so that LT4320 circuit is about 10 times more expensive than a rectifier bridge.

BTW LM393 are around 10-20 cents for 10 pieces, that is an order of magnitude even including the other needed components cost. So even accounting for the mosfet cost, Liv's circuit would be 3 times cheaper than the LT4320 one.
« Last Edit: November 20, 2018, 08:06:45 am by not1xor1 »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf