# Nominal primary voltage on transformers

Discussion in 'General Electronics Chat' started by dsp_redux, May 8, 2009.

1. ### dsp_redux Thread Starter Active Member

Apr 11, 2009
182
5
Hi, I'd like to build my first Benchtop Power Supply Unit and I'm looking at the parts I'd need. I went trough some datasheets and now I am asking myself a question. If you look at this datasheet: http://www.hammondmfg.com/pdf/5c0018-19.pdf you can see that every transformer have a 115/117V required at the primary (nominal). Why 115/117V? Historical reasons? In Canada we have a 120V/60Hz and my outlets give me 120V. I'd run my circuits with a 1.03/1.04 pu. Any idea why?

2. ### studiot AAC Fanatic!

Nov 9, 2007
5,005
519
Is it still 120 volts after a few yards of building wiring, plus the voltage drops from other devices on the outlet supply circuit?

In the UK you can loose up to 10% down a long length of building wire.

3. ### mik3 Senior Member

Feb 4, 2008
4,846
67
It is ok to run it at 120V, the difference is not very big. You will get a slightly bigger output voltage but if you use a regulator IC for the power supply then there will be no problem.

4. ### dsp_redux Thread Starter Active Member

Apr 11, 2009
182
5
I have 120V from my outlet, no voltage drop, almost perfect (nothing is perfect) sin wave (living in Quebec). I'm still wondering tough, why 115V instead of 120V? The majority of households have about 115V on their outlets?

Another question, what would be a good transformer for my power supply? It will be mostly used for alimenting chips (±5V, ±12V, ±15V) but I plan doing some audio with it too (effects, etc...). For small electronics, I don't think I will go over... let's say... 2A-3A for a big circuit? (Just guessing) How much VA should the transformer push?

$S=VI^*\rightarrow 35V\times 3A = 105W\text{ or VA if we suppose }Q=Im\left{S\right}\approx0$.

Any input is welcomed.

5. ### mik3 Senior Member

Feb 4, 2008
4,846
67
105W is the needed real power.

To find the VA rating of the transformer you need to know its power factor. Just by estimation a 120VA will be ok.

6. ### studiot AAC Fanatic!

Nov 9, 2007
5,005
519
It's easy to get too ambitious with the first of anything.
Successful power supplies require heavy duty construction, the components are not cheap.

One approach is to construct a heavy duty case/transformer/bridge/reservoir cap and leave plenty of room in the case for later additions. Getting the mechanics if this working is a major achievement in itself.

The level of regulation required by sensitive circuitry is often many orders of magnitude greater than that needed by beefy power amplifiers operating at several amps. Also lower current level regulators (less than 1 amp) are cheaply and readily available and easy to configure and install.

A psu capabale of supplying 5 volts and 30 -50 volts at several amps will be wasting lots of power, which you will have to dissipate. Better to have two supplies.

So think about a modular approach with scope for later improvement.

Good building.

7. ### dsp_redux Thread Starter Active Member

Apr 11, 2009
182
5
Yeah just estimating since $S=P+jQ \rightarrow PF=cos(\theta)=\frac{P}{S} \text{ or } S=\frac{P}{PF}$.

Q is supposed to be there since a transformer adds inductive impedance to the circuit. But I supposed it to be small or approximately equal to zero to obtain the smallest value of S needed in that case supposing it's purely resistive. Please tell me if I'm wrong. That number, 105W is just to give an idea of what I should look at. I know in fact Q is not 0.

@studiot

When you are talking about reservoir cap, do you mean by that the capacitors used the prevent the ripples from the bridge? I know it's really easy to underestimate a project like this if it's the first time... what would you suggest?

8. ### mik3 Senior Member

Feb 4, 2008
4,846
67
The transformer will have a power factor above 0.85, thus a 120VA will be fine.

9. ### dsp_redux Thread Starter Active Member

Apr 11, 2009
182
5
Will 3A be enough for a benchtop psu? I know I haven't needed more than 1A on any circuits, but those where small school projects.

10. ### mik3 Senior Member

Feb 4, 2008
4,846
67
It depends on what circuits you work with. If I was you I would build a 10Amps SMPS but this increases the cost. However, it is better if you ever need to power a high amp device.

11. ### dsp_redux Thread Starter Active Member

Apr 11, 2009
182
5
I'll need to look into SMPS. The thing is, in school we don't get to learn about power supplies but we can just push what we learned instead to build those. I could simply buy one but it'll be a more expensive choice, a less satisfying one, and you don't learn nothing from it. So, 10 amps? In what application you needed up to 10A?

12. ### mik3 Senior Member

Feb 4, 2008
4,846
67
Sometimes you need to test motors or high current coils.

If it is a school project then you have to build what they told you.

13. ### dsp_redux Thread Starter Active Member

Apr 11, 2009
182
5
No, not a school project. I will have my grad in about a year and won't have access to the school labs afterward. I'm saving money to buy myself an oscilloscope but I'd like to build my own PSU as a first "real" project. I looked at the SMPS PSUs a bit... I guess I should put a lot more effort in that than I tought, I'll read more as I get some more free time.

14. ### mik3 Senior Member

Feb 4, 2008
4,846
67
SMPS are more complex to build than a simple linear regulator.

If you want you can use a bulky transformer and a LT1038 linear voltage regulator with a big heat sink to make it.

Apr 5, 2008
16,977
2,965
16. ### loosewire AAC Fanatic!

Apr 25, 2008
1,584
438
Has any one ever measured voltage during( brown-out) over loaded
power grid. What are the tolerance on regulated power suppies. 120
down to what.

17. ### mik3 Senior Member

Feb 4, 2008
4,846
67
I think they regulate the line voltage within +-10% of the nominal value.

18. ### DonQ Active Member

May 6, 2009
320
14
A suggestion...

An abandoned PC power supply makes an excellent bench supply, and can be found for really cheap! The have lot of +5VDC current, and also have + and -12, along with a few other voltages. There are still things to do, like figure which wires are which, and supply the minimum required load (switching supplies don't work right without at least a small load on the primary output.)

Even with packaging this for your bench (terminals, meters, whatever), it will get you going long before you would get your own built from scratch. And anything more than a couple of amps should really be done with a switch-mode supply, and this is way further than you probably want to go.

You can also sometimes find supplies for things like laptops, or other electronic things, that have the output you need. Sometimes it is just easier to start with something already done and save your creative energies for things that don't have lethal voltages (120VAC) and power/heat dissipation problems that are better left for after you gain some additional skills (on much more interesting things.)

Problems with analog, digital or micro-controller circuits will do things like make the flashing light stop blinking, or make loud noises out of the speaker. Problems with high current power supplies do things like explode pass transistors, start fires, or put a buzzing sensation through your arms and chest that make you shout words that aren't nice .

I'd really rather spend my time figuring out a new micro-controller. ??

19. ### dsp_redux Thread Starter Active Member

Apr 11, 2009
182
5
I looked at the PC power supply to bench on google. Might be an option... but it removes the fun of learning new things.

On the other side, I guess I should be investing more into an oscilloscope before anything else so I can troubleshoot my uncle's blown audio amp. Any suggestion on a good oscilloscope for home debugging/designing? We worked with dual channels Tektronik OSC, but heard Agilent make some pretty good stuff too? Analog or LED display?