Do i need a second powersupply??

[Rob Sims]

Hi,

I am a mechanical engineer and I know some of the basics of electronics.
Mechanical machines and production lines are now more and more operated and controlled by means of electronics so a good understanding of electronics is necessary nowadays. I am very interested in electronics now so I'm setting up my home electronics lab. I've recently bought the UNI-T UTP305 bench powersupply. It is a great powersupply, see a review link here  https://www.youtube.com/watch?v=29P0sFswPVE. I'm  thinking about buying another powersupply, the Tekpower TP3005T. This is also a good powersupply, but the advantage of this powersupply is, that it is programmable and it has a higher resolution because it is a 4 digit supply.

My question to all the advanced electronic engineers is, do I need a second powersupply for my lab? I want to experiment with electronics (practice what I've learned in books),  test components as transistors and opamps before building them in circuits, design circuits (mostly analog I think) etc.

If I buy a dual channel powersupply I will basically have 3 powersupplies, which is overkill I think. I was also thinking about building a powersupply from a computerpowersupply.

I need some advice from you guys.

Regards,
Rob Sim

[tautech]

Welcome to the forum.

Don't do the PC PSU. 

You will at some stage need a second PSU but you can get a little creative and have your current (no pun intended) one set at the higher voltage you need and then tap off to something like a little LM317 home built adjustable to give you a second supply. The only trap is having enough voltage headroom for the 317 if the voltages required are close in value.

[MosherIV]

My question to all the advanced electronic engineers is, do I need a second powersupply for my lab?

YES!

Note sure how useful the so called higher resolution 4 digit is, only the high end precision lab PSU are any good at precision volts.
Usually, you set the output voltage while measuring with a meter because the build it meters are shockingly bad 

Generally, most labs will have at least 1 dual supply, 1 single supply and a high power (current) bench supply.

PC PSUs are good for the high current BUT (and this is a BIG 'but' for lab work) they do not have a current display or current limit.
So they can destroy delicate electronics before you notice, but if you cannot afford a high power supply they can do in a pinch.

[Rob Sims]

Many thanks for the advice guys. I will  buy the Tekpower, cause I don't want to start too heavy.
My next buy will be a Rigol oscilloscope.

Regards,
Rob

[Dave]

Whether or not you need more than 1 power supply depends entirely on what you are doing. If you want to test circuits that require more than 1 voltage rail, having 2 or more power supplies does come in quite handy.

I would suggest that you upgrade once you actually feel the need for an additional power supply, buying it for no good reason is rather silly, because its specs might not meet the requirements of your next project.

[Mechatrommer]

Don't do the PC PSU. 

why not?

[MosherIV]

Don't do the PC PSU. 

why not?

Computer PSUs do not have the ability to set a current limit,
They do not have a display of the current draw.
can be very noisy dur to switching
Do not normally have variable voltage
 
To name but a few problems.

However, they are a cheap way to get a high current PSU when on a tight budget.

So, it is best not too if you can afford a proper one

[Mechatrommer]

Don't do the PC PSU. 
why not?

Computer PSUs do not have the ability to set a current limit,

i mean, instead of the norm transformer+rectifier... why not use PC PSU as power source for a homebrew linear regulator with current limit circuit?

[MosherIV]

Noise. Computers are actually quite tolerant of switching noise, so the PSUs are actually quite noisy. It helps keep the costs down.

By the time you add enough filtering on, the cost and decrease in efficiency it becomes less and less worth while.
You can never truely get rid of the switching noise so most low power bench PSUs are linear.
Many uses for bench PSU require very clean voltage eg op-amp circuits.

Pc PSUs are very cheap and many people do use them but if you can afford it, it is better to go for a proper bench PSU

[mikerj]

Noise. Computers are actually quite tolerant of switching noise, so the PSUs are actually quite noisy. It helps keep the costs down.

You also have quite a limited supply range, 12v less whatever the drop-out of your regulator is.

[rstofer]

The way I see it is that a triple output supply is the way to go.  Conventional op amp circuits need +-15V and the following circuitry may need +5V.  I don't have such a triple output supply because I spend most of my time worrying about microcontrollers.  I am starting to mess around with op amps so I will probably buy the Rigol DP832.  I am hoping that the volt and current meters are accurate because many experiments in Learning The Art Of Electronics want both parameters measured.  I will check them with a DVM, of course.

[Rob Sims]

Thanks for the advice guys.   I'm considering the cheap Tekpower TP3005T as a second powersupply because:

- In case  one powersupply fails, I have another one ready. If I buy a dual or triple channel expensive brand i'm stuck completely if the unit or
   part of it fails.
- I think for testing opamps etc you need more than one powersupply.

I want to do research, build circuits, explore other fields than testing motors only etc. I want to grow. Can you name some more circuits or work that require two or more powersupplies?

Thanks

[Mechatrommer]

Noise.

but but... we do have options for high powered yet lightweight bench psu dont we? like this one.... http://www.mcmelectronics.com/product/TENMA-72-7655-/72-7655 20-60A capable is not less than 2kg in linear world, and not cheap...

[MosherIV]

but but... we do have options for high powered yet lightweight bench psu dont we? like this one.... http://www.mcmelectronics.com/product/TENMA-72-7655-/72-7655 20-60A capable is not less than 2kg in linear world, and not cheap...

Yes, and you quite right, these are probably switch mode PSUs.

If you measure the output with an oscilloscope, you will probably find the switching noise is there, probably up in the 100KHz region and probably no more than 2mVp-p

Linear bench PSUs tend to be lower power (upto 30V and no more than 4 or 5A). Beyond this, they tend to be SMPS type.
Most lab/development work that needs high(er) power generally does not need the absolute clean voltage eg motor driver applications

[Galenbo]

-Computer PSUs do not have the ability to set a current limit,
-They do not have a display of the current draw.
-However, they are a cheap way to get a high current PSU when on a tight budget.

True, but the creative mechanical-electrical hobbyist can do some easy mods:

-add a $2 dollar V+A Led Display.
-with a switch, select to add a series resistor. (with the right Watts spec, 10ohm/20W and/or 5ohm/50W )

Explanation for beginners: the 10ohm series resistor will give you still a good 11,8 V @ 20mA, so OK for led experiments, and limit to only 1V if you accidentially short-circuit or something suddenly needs 1,1 A.

Connect the display after this series resistance, so you can see what happens with the voltage.