Jump to content
Electronics-Lab.com Community

My first project - A simple regulated power supply 30V


Guest Jaxelo

Recommended Posts

Hey,
I've been getting into electronics lately (that's what I study, but we've been doing theory more or less so far) so I decided to build a simple regulated voltage supply that can go to about 30V (doesn't really need to go down to 0V). I've read the big thread but decided it's probably better to start small. I've seen this simple circuit and I can understand how it works (well, not the chip itself, but I understand its function/role in the circuit). I decided to post it up here so I can get some feedback from the more experienced guys and let you know of my progress as I go. Also, I might make it be able to be supplied through a transformer from mains or from a phone or some other similar device's charger (DC).
So for starters I'd like to know your opinion on this schematic. I'm open to modifications (keep in mind this most probably won't be used to power something sensitive/precision stuff).

Link to comment
Share on other sites


Guest liquibyte

LM317's are ok and will power stuff nicely and forever if you stay towards the upper limit between input and output but just be aware that as you lower the voltage the device itself has to dissipate the wasted energy in heat.  What I'm saying is that this will require serious heatsinking for the most part if you're inputting say 25V and want to power something at 3.3V.  C3 doesn't need to be that large a value.  I suggest you study the datasheet for the part to get a better idea of its limitations.  I keep several around for various reasons and started out using them to learn with.  You'll find out soon enough that they don't quite fit the bill sometimes.  Don't be intimidated by the big thread and the designs shown.  It's a very good supply for what it's worth and we'll eventually work out that last bug.

Link to comment
Share on other sites

Guest liquibyte

I think you'd have a hard time trying to find that transformer without having it custom wound.  You'll have that issue with many of the commercially based designs anyway.  Off the shelf odd values just aren't readily available without a salvage mission or paying for one-offs from the few places that will do that kind of thing.  I've checked into it and it can cost quite a bit.  Winding your own is another option but be prepared to learn quite a bit of transformer theory before you start.  If you want to explore various power supply designs, head over to KO4BB and BAMA and start downloading service manuals.  HP/Agilent and Tektronix have some excellent stuff in that area.  The newer models will have quite a bit more complexity but you'll see some solid designs from the 70's and 80's that aren't too hard to work out.  We figured out the shutdown transient issue with the 0-30V version here needed Q1 put back in based on a discussion I was having about the same type of circuit in the Tektronix 503a.

Now that I've said all of that, a basic 317 design is ok to start with but won't get you where you'll eventually need to be power wise for your future projects later on down the road.  Since I'm involved in the big 0-30V project, I intend on following the thing through to completion but I'd suggest that for anyone getting into this as a beginner would be to look for a nice 0-20V 0-2A dual design either as a project from the aforementioned service manuals, another project on the net, or as an Ebay find as something to start with.  The reason I say this is so that when working with op amp projects you'll have a solid power supply to use, and you'll eventually be using quite a few op amps down the line and will need power for the dual supply types sooner or later, trust me.  Once I finish the one I'm working on I plan on doing such a project myself as a dual tracking design that will probably be based around one of the older Tek designs.

Link to comment
Share on other sites

I've got the components for the 317 circuit so I may actually be putting it together tonight when I get home.
About the "real" psu I was looking at this; it does require an odd transformer but I think I can manage + I could run something like an ATTinny on the +5 rail.

http://www.elv-downloads.de/Assets/Produkte/2/225/22532/Downloads/22532_Universelle_Netzteilplatine_um.pdf

Link to comment
Share on other sites

Guest liquibyte

It's not really necessary but is a nice thing to have.  I'd have never found the transient issues had I not actively been looking for them using mine.  Well, I did see things that were odd using my meter but I just chalked that up to a cheap meter.  When I started comparing results from both devices is when I noticed something very odd going on.

Make sure that when you build this circuit to use the biggest heatsink you can get for the 317 and possibly a fan for active cooling as well.  They have thermal shutdown but keeping the device well away from that point is always a good thing and will ensure the power supply lasts.  As a general hobby supply, I'd recommend keeping the protection diodes in there even though you can find versions online that don't have them.

Link to comment
Share on other sites

Yeah, I know - used them in some courses and the digital ones are really nice :D

I did some calculations and understand that. This is my first circuit so I thought best start small/simple. Next thing I'll be building is probably that german thing since I wanna have a nice PSU but it's such a simple thing I don't want to buy one (although it may be cheaper).I talked to some people and a few said I should get into the switching type of supplies. No doubt I will sooner or later have to get to know those but I don't mind a dual function supply (especially in the winter :D)

My next step is laying out stuff on a stripboard, but I won't be home till next weekend so I won't be able to get to building it

Link to comment
Share on other sites

So i put this thing together on a breadboard using a 9V battery as my input just for testing. Voltage is 8.67 V and I get 1.25 - 7.65 out. (numbers are a mere approximation, hence the multimeter)
I don't have a transformer yet and I think there's no point in putting this into a case for use - I wanted to build something simple first to see if I can get it to work.

I will probably be going with that ELV design, but I'm gonna try to talk to some people at my uni for their opinions first - of course I'd want you guys' opinions as well (that's why I'm posting this on here :D).
I want to have a 2 x 15V transformer so I can switch between 15/30V with a relay - need to design that part of the circuit but it shouldn't be too problematic. Could also be done with a mcu if I end up putting one in (5V op-amp supply), which would be used for displaying voltage/current.

post-114711-14279144825938_thumb.jpg

Link to comment
Share on other sites


..............I don't have a transformer yet and I think there's no point in putting this into a case for use - I wanted to build something simple first to see if I can get it to work.

I will probably be going with that ELV design, but I'm gonna try to talk to some people at my uni for their opinions first - of course I'd want you guys' opinions as well (that's why I'm posting this on here :D)...............................

If it’s not too late to change direction, have you considered going down the switch-mode route?

My ageing 28V 2A power supply (using 2N3055) is now getting past its ‘sell-by’ date, so I intend to make a new one using ready made inexpensive modules from ebay.

I already have an HP laptop power supply that delivers 18.5V at 3.5A. That will power a buck converter to give an adjustable output of 1.8V to 16V at up to 5A (subject to the 65W maximum of the power supply).  The buck converter module has two presets to adjust the output voltage and the current limit. I shall replace these with normal pots. The output voltage and current will be displayed on two 3digit LED displays – again using ready-made modules.

To get a higher voltage, when needed, I shall switch in (between the power supply and the buck converter) a boost converter that can give up to 32V output at up to 3A. That will then give me an output of 1.8V to 30V.

I’m waiting for the displays to arrive at the moment, then I’ll start hooking it together. If you are interested in making something similar, I can give you the ebay item numbers.
Link to comment
Share on other sites

  • 3 weeks later...

I've been working on that ELV psu - I haven't actually built anything but I ran lots of simulations and I solved some problems: The old voltage regulation was kind of strange - it was letting too much voltage in and the regulation wasn't even working at one end of the potentiometer. I also changed the pot from 100k to 10k and will probably later be swapping that with 2 pots for coarse/fine adjustment.
One other thing I did was remove the max voltage out set pot since I think I won't really need it.

I think all of these simulations might be a bit inaccurate since I am using a lm7905 model that actually outputs -5.3V. Could that effect my measurements in a way that what I am seeing here is not what I would see if I built this up?

I'm providing some pics (simulation and schematics)

Edit: first two pics are of the changed circuit, the other two are the original circuit in the pdf from ELV

post-114711-14279144874755_thumb.jpg

post-114711-14279144875228_thumb.jpg

post-114711-14279144875551_thumb.jpg

post-114711-14279144875821_thumb.jpg

Link to comment
Share on other sites

Guest liquibyte

There is this 0-30v power supply. It's okay. Opamps need a higher power supply voltage because of the complexity of the circuit.

Do you get some perverse pleasure out of posting crap like this or are you just drunk all the time?
Link to comment
Share on other sites

Guest liquibyte

The output is at the LOAD resistor and the schematic has a similar layout to the original one if that helps

Is it supposed to be a negative supply then because that's what I'm getting?

post-107142-14279144876141_thumb.png

Link to comment
Share on other sites

Guest liquibyte

When it comes to electricity and simulation it doesn't matter what point your ground is referenced at but it makes it hard to get a positive voltage in the sim.  Yeah, I duplicated it from the image.  The folks responsible for LTSpice are wonderfully prescient about keyboard usability so I'm getting quicker as I train my habits that way and enjoy doing it.  I wish there were a translated version of that pdf though, I'd like to read the article.

I meant to ask you if you knew how to add parts from models off the net?  You can actually put the LM324's (or any op amp model you have) in the circuit if you have the spice file for it using opamp and opamp2.

Link to comment
Share on other sites

The + rail is grounded for regulation purposes as I understand. Could I put a "real" earth (from transformer) on the negative rail if I wanted to reference it to ground? Additional green plug would be used for that.
Also the thing I put in instead of the max U set pot might not be optimised, not sure about that.

I also like using the key shortcuts but am not that familiar with the software... I don't really understand how the model/library thing works with LTSpice... do you?

Link to comment
Share on other sites

Guest liquibyte

The + rail is grounded for regulation purposes as I understand. Could I put a "real" earth (from transformer) on the negative rail if I wanted to reference it to ground? Additional green plug would be used for that.
Also the thing I put in instead of the max U set pot might not be optimised, not sure about that.

I also like using the key shortcuts but am not that familiar with the software... I don't really understand how the model/library thing works with LTSpice... do you?

Yep, I do know how most of this program works but there are certain aspects of spice and LTSpice that go way over my head.  All you have to do is be dedicated and do your research and you'll find ways to do things that you didn't think possible for a simulation.

Here's a model I did based off the pdf that has the LM324, TIP142's, and I used a resistor with a temperature variable as the PTC.  I'm still trying to figure out where the startup transient is coming in at but the model is valid.  Read the README because you have to put the stuff for the TIP142's inside your programs folders so they work and you can call them up with your F2 key.

post-107142-14279144876482_thumb.png

ps.zip

Link to comment
Share on other sites

Cool, I need to get a better understanding of the program :D
How do you link the +-5V down to the opamps?

Btw what's the output of your lm7905? I have a model that outputs -5.3V

Also - this thing is supposed to go 0-30V but it couldn't ever do that with a 30V transformer, so I'm probably gonna make it go to 25V or use a 34V transformer or something like that.

Link to comment
Share on other sites

Guest liquibyte

Cool, I need to get a better understanding of the program :D
How do you link the +-5V down to the opamps?

F4 > Label net > Choose output for the output voltage (i.e. the net at the output of the LM7805) you want input at the op amps and attach that label there.  Then do the same for the input of the op amps but choose input and name the label the same thing you did at the output.  To summarize, you're naming the net for the output voltage at the voltage regulators +5V and -5V and the net at the input of the op amps +5V and -5V.


Btw what's the output of your lm7905? I have a model that outputs -5.3V

-5.3086687V


Also - this thing is supposed to go 0-30V but it couldn't ever do that with a 30V transformer, so I'm probably gonna make it go to 25V or use a 34V transformer or something like that.

I get -25.357171V from ST9 to GND.  I'm still playing around with it to see if I can stabilize the input and I'll have to translate the pdf to see if I can understand the circuit a little better to make it output what It's supposed to be doing.  Changing the input voltage to 34-35V won't change the output so the issue lies elsewhere.

Have you tried breadboarding this yet?  I'm curious because it doesn't simulate very well out of the box.
Link to comment
Share on other sites

Guest liquibyte

I've been looking through the kits they offer there and I don't think any of it is really very good from a truly innovative standpoint.  Perhaps we should be trying to take some of the old power supply service manuals from HP and Tektronix and working those up into workable designs.

Link to comment
Share on other sites

Thanks for the info!
I can get steady 30V if I put 34 sine input - I think only the voltage divider that feeds into the opamp needs correction. I talked to a guy at our university which is "specialized" in electronics and he said (after a short inspection) it should be fine. There might be an error with the resistors, though.

I've been thinking about that lately as well - might be the best option but we need to find something not too complicated/big.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
  • Create New...