Design Power Supply 30V / 5A

[Ahmed Khouaja]

Hi Everyone

I'm new member here in the EEVBlOG, i really want some help to design my own lab power supply

The Specifications required in my lab power supply are  :

• Adjustble voltage from 0 up to 30 V
• Adjustble Current from 0 up to 5 Amperes
• Inversion protection based on relays
• Possibility of current injection while short circuit is found ( inject current in laptop motherboard to define the deffected composant)

i hope that you can help me guys and thnx a lot .

[tooki]

68 people read this without replying, which suggests the question is poorly formed. We aren’t going to do the design work for you. So why don’t you instead post what concrete problems you’re running into.

[Ahmed Khouaja]

Sorry guys but i meant any proposition to give for my project because i still in the theory part ...

[tooki]

Nope, that part is entirely up to you!!!

[Ahmed Khouaja]

Okey i'll post everything i found and we will check if its right or not. Thank you

[Ian.M]

If you build a pure linear supply with those specs its going to be an expensive large multi-Kg brute, as the pass transistors have to dissipate around 190W if its suddenly shorted while driving 5A at 30V (assuming 38V average raw in allowing for 2V of headroom and the rest ripple).   Yes there are things you can do to mitigate that, like tap-changing or switched mode preregulation, but at the end of the day, your output stage has to be able to handle the peak power long enough for the tap changer or preregulator to react, and all the mitigation strategies compromise the voltage slew rate when the load impedance suddenly increases in constant current mode.   Its hard enough to get right for a 24V 500mA rated supply, so for a 30V 5A supply we'll be sitting around with  while you find out how many expensive transistors  you can let the magic smoke out of per day.  I predict you'll fill at least one 1lb jam jar with them in a month of full time lab work, and that's not counting the ones that depart this life as silicon shrapnel.  Hopefully you are fit enough not to get a hernia lifting the brute!

If you want to do a SMPSU, that has its own set of challenges that aren't helped by working with 150W of output power capability, and if you have the skills to tackle that project, you wouldn't be asking novice questions here, so I predict even more  for your audience, MOSFETs in the jar instead of BJTs and a high enough incidence of silicon shrapnel that  you couldn't pay me to work at the the same lab bench.  The only plus point is you wont get a hernia lifting the PSU chassis on and off your bench.

Of couse you could just order a Hangzhou Ruideng Technologies DPH3205 buck/boost digital PSU module from China, then all you've got to do is hook it up to a good quality second hand 180W laptop charger, but where's the fun in that?  There certaily wont be a lot of entertainnent for your audience unless you do something really 

[lowimpedance]

"silicon shrapnel"         

     not heard that before, thanks for the mental image created !.
 

[Ian.M]

Well what else do you call it when the whole of the top of the package and the die have attempted to go into orbit and all that's left is the tab, some part of a badly fractured package bottom more or less stuck to the heatsink and the twisted and partially vaporised remains of the lead frame?

[lowimpedance]

Well what else do you call it when the whole of the top of the package and the die have attempted to go into orbit and all that's left is the tab, some part of a badly fractured package bottom more or less stuck to the heatsink and the twisted and partially vaporised remains of the lead frame?

Indeed ! , luckily for me I've spent a sheltered life in low power with a more 'generally' smaller flame and smoke combo ......and less shrapnel.

[salbayeng]

Well tantalum shrapnel is worse, as it's incendiary, and pongs for days.

Another one went somewhere in my RFI spectrum analyser about a month ago, and I haven't got the time to find it, the last one took me a day or so to get the 1000 odd screws out to find it , after replacing that smoking one, I could hear a rattle in one of the brass boxes, so I unscrewed 2 boxes (another thousand screws) before I found a whole pile of shrapnel inside a box, took about 5 minutes to find where the tantalum had been, there was literally only 2 1mm stumps left of the leads.

[JS]

Please start with something smaller!

You don't need much to track a problem in a mobo, 1A is plenty, two LM317, 9V transformer and will fit that purpose.

Why you need such big power? is there any project you are working on right now that requires it? I've built myself a dual supply capable of 24V and at least 5A (never tested it) but used once or twice, not lab or current limited, was designed to use in a project as ±24V supply, so heatsinks aren't close to big enough to supply the 5A at any low voltage, but still comes handy every once in a while. I'm getting a proper one this weekend, much smaller but quite useful in the lab.

JS

[Kleinstein]

For learning it really makes sense to start smaller, though this does not mean using LM317.  One can still use the concept with separate power transistors at a low power (e.g. 20 V and 1 A). It's only after one has build, tested and understood this smaller one, that one might go for a larger one.  Keeping the voltage a little lower (e.g. 20 V) helps because one could use more normal OPs without exceeding there voltage rating.

[xavier60]

Because OP intends to use the power supply for finding Main Board shorts, the chosen design must be able to adjust down to 0v which some designs don't. Like the LM317 based ones.
I also wonder why it is necessary to be able to supply up to 30v when 19v is the most that I have seen needed for computer work.

[KrudyZ]

Hi Everyone

I'm new member here in the EEVBlOG, i really want some help to design my own lab power supply

The Specifications required in my lab power supply are  :

• Adjustble voltage from 0 up to 30 V
• Adjustble Current from 0 up to 5 Amperes
• Inversion protection based on relays
• Possibility of current injection while short circuit is found ( inject current in laptop motherboard to define the deffected composant)

i hope that you can help me guys and thnx a lot .

Like others have said, this might be a bit too much to chew off if you are a beginner.
I would suggest buying a Basetech BT-305 supply which has the exact specs you quoted.
It is a linear regulator with multi-tap transformer to adjust the drop on the pass transistors, can to CV CC with both an AC switch and output enable via relays.
The full schematics are easily available, in case you really need to make a modification.
It will be a lot cheaper to buyas a unit, than buying the parts in small volume.
If you really want to build this yourself, then still buy the unit, take it apart and put it back together

[bloguetronica]

Hi,

I think your biggest challenge will be having voltage control and adjustable current limiting at the same time. That is not easy to do, at least efficiently (that is, using a minimum number of parts). Note that these two factors, as per the specification  you presented, must be intertwined. You are monitoring both voltage and current at the same time, and your supply must decrease the voltage accordingly once it gets to the constant current regime, in a precise manner.

You can simplify your design significantly if you simply use a fixed short circuit protection. Really, I suggest that you start simple. Play with simple pass transistors with Zener references, then add an op-amp and see what happens. You can learn so much by experimenting this on a breadboard, considering that you downsize your specs significantly (here goes the 5A and the 30V). Then add a basic fixed short circuit protection. Then you can implement a sort of pre-regulator DC-DC converter to your circuit if you want more range in the voltage and current department.

Take into account that those 30V and 5V lab supplies that you see often use multiple pass elements (pass transistors) in parallel, and then you have to guarantee that the current is evenly distributed, or you'll exceed SOA with catastrophic results. Then take into consideration the intricacies of how the CV/CC control is achieved. For instance, I have a schematic of my PSU and all I can say is that it is very complex. Also, the mains transformer has multiple taps in the secondary, to further guarantee the SOA of the pass transistors.

I hope that, by no means, this passes through as being demoralizing. It is not my intention. It is better to start simple and then do more complex things further down the road. You'll have a valuable insight and learn way more.

Kind regards, Samuel Lourenço

[Ian.M]

Well what else do you call it when the whole of the top of the package and the die have attempted to go into orbit and all that's left is the tab, some part of a badly fractured package bottom more or less stuck to the heatsink and the twisted and partially vaporised remains of the lead frame?

Indeed ! , luckily for me I've spent a sheltered life in low power with a more 'generally' smaller flame and smoke combo ......and less shrapnel.

At least the TO-3 metal can of the near-obsolete 2N3055 that seems to still be popular for poorly designed high current linear PSUs can be counted on to contain the *BANG* so no silicon shrapnel, though I did see a TO-3 once that had managed to eject its base pin, which was left flopping around on the end of its hookup lead complete with the remains of its hermetic seal!

@KrudyZ: Thanks for the lead on the Basetech BT-305.  I'm tempted to buy one as a linear PSU development testbed, or even general bench use.

[Ahmed Khouaja]

Sir , i think you have the right , so i changed my plan , i will use simple lm317 regulator with some pass transistor to amplifie the current otherwise i've seen a good video for a current protection it could help me out and an inversion polarity protection that i mentionned in the previous replay. thank you all for your help. and i will post the project after i finished him.

[Ian.M]

AAGH!!! *IF* you *MUST* use an old regulator and add an external pass transistor, at least *LOOK* at the LM723 before considering a LM317.   The LM723 is designed to support an external pass transistor with minimum extra parts, and includes basic current limiting that only needs one low-ohm resistor to set the current

[Ahmed Khouaja]

 The probleme is not in the LM723 itself , but in our electronics stores , may i cant find it so i need to use lm317 because i already have it ...

[mariush]

here's schematic for such psu (you can simplify it for ex. by removing adjustment using push buttons and ignoring the voltage and current displays etc)

[xavier60]

I like the transformer secondary winding switch idea in the Dick Smith PSU.
I'm not certain if it's wise for the OP to jump directly into a MOSFET design, this is the circuit I used lately, https://www.eevblog.com/forum/projects/lab-power-supply-the-lost-current/?action=dlattach;attach=489698;image
The large MOSFETs can be found very cheaply, I have just ordered some PSMN009-100W from Rockby for 90 cents each.
In my application the circuit can supply 20 amps but is fed from a pre-regulator. It should safely handle 5 amps, with adequate heat sinking, if fed from the same transformer setup used in the Dick Smith PSU.

[bloguetronica]

here's schematic for such psu (you can simplify it for ex. by removing adjustment using push buttons and ignoring the voltage and current displays etc)

Hi mariush,

The third schematic looks like the PSU I have. There are a quite a few differences, though. Never delved in the details on how the current limiting works. IMHO, I can't spot the current sensing. These designs are unintelligible for me, and that's my main difficulty. Curiously, enough, mine uses LM714 op-amps.

I've attached the schematics. The manual has a schematic for a two pass transistor supply as well. The LED voltmeter and the LED ammeter seem to be intelligible enough for me.

Kind regards, Samuel Lourenço

[salbayeng]

OK Samuel,   for the page 3 schematic, it's pretty simple , start our journey with the two LEDs V15 and V16.

-------- current control mode -----
if V15 is illuminated, then you are in current mode, current through V15 drags down the drive to the  pass transistors (v13,14,21,22). So how does it get into current mode? simple!, the voltage across the current sense resistor R10 exceeds the voltage across the current pot RP3.  The difference in voltage is sensed by N2 (a garden variety LM741).

--------- voltage control mode -----
if v16 is illuminated, then you are in voltage mode,  this happens if the output voltage is greater than the voltage (at C13) on the wiper of the voltage setting pot RP4 , RP2. , The difference in voltage is sensed by N1 (a garden variety LM741).

--------- floating bias supplies ----
What make the whole circuit confusing is the "gnd" of the circuit is actually the positive output terminal,  and there are floating voltages referenced to this node,  e.g. the TL431 generates 2.5000 V reference above the positive output terminal ,  there is +12v on C3 , and ~ -15v on C2 .

-------- tap changing ----
There are two relays K1 and K2 that change transformer taps according to the output voltage.  M4A and M4C  compare the output voltage with two setpoints one across C15 , the other across C17 , these two caps slow down the voltage levels so the relays don't switch unnecessarily with small voltage dips.  R29 and R28 provide hysteresis so the realys don't chatter back and forth if you are close to the switching threshold.  Note that the relays perform as primitive analog to digital convertors,  so K1 adds say 5v to the transformer voltage, K2 adds 10v, and both add 15v.  To get the A-to-D effect , V28 shorts out R23 , which adds 10v to the reference voltage of the 5v relay driver.

----------- stabilities  --------
Almost all power supplies are basically unstable, as you have two or three lag terms in the feedback loop , there is a really good tech-note out there written decades ago that explains all this, I can't find it, but this isn't too bad, from about page 5 , this is the topology commonly used. http://www.ti.com/lit/wp/snoa842/snoa842.pdf  go ogle "voltage regulator dominant pole"
In any case the best solution to the stability prooblem is to use a SINGLE DOMINANT POLE , i.e basically make one part of the circuit so slow, that it swamps the delays in other circuit elements, in your circuit  it's C7 in voltage mode, and C10 in current mode (you then need to add C8 and C11 to "compensate" the LM741 , because it is prone to oscillate internally with low local loop gain)
If you design your own PSU , you will also need to add a dominant pole somewhere, the mistake most newbies make is to have too much  gain in the loop , you only need unity closed loop gain to make a PSU.

------------------- pg4 ---------------
Page 4 looks like a higher current (and maybe higher voltage) version of the same PSU , it adds an extra pass transistor , and an extra voltage tap ,  so the relay driver is now a 3 bit A-to-D, so you might have
       taps at 5,   10,    15,   20, 25,  30,   35, 40v 
for K3,2,1 = 000 001 010 011 100 101 110 111
(I'm just guessing these voltages BTW)

--------------------- fan --------
Most PSU's have a thermally activated fan on the heatsink

-------------------- accuracy -------
All the components used are very cheap, garden variety, so you can expect the last digit of a 4 digit LED display for volts and amps to be fairly meaningless

[bloguetronica]

Thanks salbayeng,

Your post was very insightful with lots of information. I'll have to study this to get some ideas.

Kind regards, Samuel Lourenço