Help with cryptic UA723 powersupply schematic

Thread Starter

ChrisChemist116

Joined Mar 13, 2009
79
Hi there:), after looking for several schematics to design a power supply which can fit to my needs i found the following model.

The original file powersupply723_original.PNG has been taken from this site: http://electroschematics.com/210/adjustable-power-supply-0-30v-1a/comment-page-1/#comment-8359)

I like it because it doesn't exhibit rare/ exotic or ludicrous to find components.

However because the scanned image it has low resolution i have practically no any idea which nodes are in junction and which aren't.
Could anyone on this forum give me a hand on this?:confused:

I did my best on putting my questions in the schematic, be easy on me okay? (my question is on powersupply723_question.PNG)

I dont know where did this schematic was taken from, but my intuition tells me that maybe comes from an old 300x Elektor magazine circuits compilation of summer circuits from the 80s-90s, maybe...but i am not sure.

My current knowledge in electronics is just beginner, but i understand many people here has a diverse range-expertise in electronics, and i believe anyone out there could suggest me what to do to solve my problem.

Thanks:rolleyes:
 

Attachments

Thread Starter

ChrisChemist116

Joined Mar 13, 2009
79
Hello,

The crosses with the dots are connected.
The crosses without dot are NOT connected.

See also here in the eBook:
http://www.allaboutcircuits.com/vol_5/chpt_9/1.html

Greetings,
Bertus
Okay i already knew that, but my question was refering on that circuit.

I can't tell which ones are connected and which arent because the image it isn't clear. The dots in the powerline coming from the bottom of the diode bridge don't show the junction very well.:confused:

Is there any suggestion or hint that can gave me a clue which nodes should be connected? i just dont want to invoke any accidental short circuit or meltdown the 723.

Thanks anyways :cool:
 

Thread Starter

ChrisChemist116

Joined Mar 13, 2009
79
Have a look at the attached.

Does that make more sense to you?
Thanks Sarge:) for redrawing the circuit. It really shows things clearly.

However i still have some questions...

In the redrawn circuit it is shown the 2.2K resistor is connected in series with P1 10k potentiometer.

However, do i understand that the cross section between these two doesn't imply there is conection with the two zeners on the left side?.
What would happen if there is a conection?. Short circuit?.

How about the 7 pinout (V-) coming from the 723. Does that mean that negative voltage comes from the output of the transformer passing through 1N4001 diode?

It seems P2 (10k potentiometer) and R6 1K resistor are in series but not connected to the powerline in the bottom.

Why is that? is it because makes a relative ground coming from the diode bridge?.

Thanks:rolleyes:
 

Attachments

SgtWookie

Joined Jul 17, 2007
22,230
Thanks Sarge:) for redrawing the circuit. It really shows things clearly.

However i still have some questions...

In the redrawn circuit it is shown the 2.2K resistor is connected in series with P1 10k potentiometer.

However, do i understand that the cross section between these two doesn't imply there is conection with the two zeners on the left side?.
Two wires cross, but there is no dot. That means that they are not connected. If they WERE connected, you would see a dot.
What would happen if there is a conection?. Short circuit?.
The circuit would not work properly.

How about the 7 pinout (V-) coming from the 723. Does that mean that negative voltage comes from the output of the transformer passing through 1N4001 diode?
C1/D1/D2 is a charge pump.

Why don't you simulate the circuit in Spice so that you understand it better?

It seems P2 (10k potentiometer) and R6 1K resistor are in series but not connected to the powerline in the bottom.
Is there a dot? No. R6 is not connected to the horizontal wire immediately below it. However, it IS connected to the wire that connects to the IC1 pin 7.

Why is that? is it because makes a relative ground coming from the diode bridge?.
You really need to start simulating things and building things so that you understand them better.

Either that, or start off with a less complex project. This one might be a bit tough for someone with little-to-no experience in electronics. Taking on projects that are beyond one's understanding can be quite risky.
 

Thread Starter

ChrisChemist116

Joined Mar 13, 2009
79
Two wires cross, but there is no dot. That means that they are not connected. If they WERE connected, you would see a dot.

The circuit would not work properly.


C1/D1/D2 is a charge pump.

Why don't you simulate the circuit in Spice so that you understand it better?


Is there a dot? No. R6 is not connected to the horizontal wire immediately below it. However, it IS connected to the wire that connects to the IC1 pin 7.



You really need to start simulating things and building things so that you understand them better.

Either that, or start off with a less complex project. This one might be a bit tough for someone with little-to-no experience in electronics. Taking on projects that are beyond one's understanding can be quite risky.

Okay i will do a simulation of the circuit.

However i want to make myself clear that, i wasn't sure which wires should be connected and which shouldn't, but not because of misunderstanding of the schematics convention, but rather the author's logic involved with the design of the circuit.

I already have made similar projects with almost no any problems at all (breadboard). So i personally dont believe this is complex to me. :)

I do have knowledge in electronics but as you noted is little. And for me, at my current level, the mix between old and new conventions in notation can endear a risk sometimes when is not seen clearly in the schematic.

Personally is believe there is no harm to make questions even if they seem dumb or iterative, however i believe little experience can easily trigger a lot of questions that can eventually lose patience to anyone with full expertise, if that happened i'm sorry.:(

I do have Multisim 11 on my computer and i think i can test it with suggestions noted.

However that arises the question on what software should be used to make a proper spice simulation of this circuit properly.?:rolleyes:
 

SgtWookie

Joined Jul 17, 2007
22,230
I don't have Multisim, so I have no input for you about it.

I've been using LTSpice and another old and obsolete simulator. LTSpice does not have a 723 regulator IC in the default library. I found a model and downloaded it, but have not tested it as of yet. My time is just getting to be a bit too limited nowadays.
 

SgtWookie

Joined Jul 17, 2007
22,230
I threw together a simulation in LTSpice, and it seems to work for the few tests I ran it.

Pin 7 of the 723 is getting -5v; that's due to the charge pump action of C1/D1/D2 as I mentioned before. R1 knocks out any ripple on the -5v, and limits current through the 4.7v Zener. D3 clamps the max 723 supply to 33v.

Anyway, even if you installed LTSpice, you wouldn't have all of the models available, and I simply don't have the time today in order to get you up and going. Suffice to say that it works in simulation, built right from the schematic you posted (and that I added the dots to).
 

Thread Starter

ChrisChemist116

Joined Mar 13, 2009
79
I don't have Multisim, so I have no input for you about it.

I've been using LTSpice and another old and obsolete simulator. LTSpice does not have a 723 regulator IC in the default library. I found a model and downloaded it, but have not tested it as of yet. My time is just getting to be a bit too limited nowadays.
Okay i understand about "time", it is also getting limited to me as well, so no worry about it. You've been nice already on assisting to me on this issue.

I heard about LTSpice but i haven't installed on my computer yet. I will try it.;)

I threw together a simulation in LTSpice, and it seems to work for the few tests I ran it.

Pin 7 of the 723 is getting -5v; that's due to the charge pump action of C1/D1/D2 as I mentioned before. R1 knocks out any ripple on the -5v, and limits current through the 4.7v Zener. D3 clamps the max 723 supply to 33v.
I did some review on my book about charge pump, and now i understand very well about what you say.

Anyway, even if you installed LTSpice, you wouldn't have all of the models available, and I simply don't have the time today in order to get you up and going. Suffice to say that it works in simulation, built right from the schematic you posted (and that I added the dots to).
Umm.. okay Sarge, i will see what i can do to find models for LTSpice. However i have tested it from your schematic on (National Instrument's Multisim11) and it seems to work fine as well.
I will test it in a breadboard later to see how it behaves. :cool:

Cheers:)
 

SgtWookie

Joined Jul 17, 2007
22,230
I'm attaching a couple of models for you to use with LTSpice:
LM723.zip - contains:
LM723.sub - place in SwitcherCad\lib\sub
LM723.sym - place in SwitcherCad\lib\sym\PowerProducts; that way it will show up in the list with the Linear Technology regulators.

pot.zip - a potentiometer model. It contains:
pot.sub - place in SwitcherCad\lib\sub
pot.sym - place in SwitcherCad\lib\sym

LTSpice does not have suitable Zener diodes in the default library. I have added some to my custom library, but that's a project for another day.
 

Attachments

Thread Starter

ChrisChemist116

Joined Mar 13, 2009
79
Its been several weeks already since i posted this message. The thing is i did finally managed to get all the components and i mounted the circuit as noted in my breadboard.

After several atempts everything failed. Actually when i measured the voltage coming from pin 3 from the LM723 to the GND (coming from the rectifier bridge) and i moved the potentiometer (P2) nothing happened. The voltage just moved from 10volts up to 30 and so back and forth.

I tried to reduce or adjust to 0V by adjusting the potentiometer (P1) coming from pins 5 to 6 on the LM723, once again nothing happened.
I believe the design on this circuit is wrong, or at least the results i got it is not something i was looking for.

Well i tried to do the simulation, but it didn't work as expected neither. Anyways it is always good to try new tools such as LTSpice as it is free.

Any help?:confused:
 

SgtWookie

Joined Jul 17, 2007
22,230
OK, I'm attaching the LTSpice simulation I did awhile back.

In order to run this simulation, you will first need to have LTSpice installed on your computer. Google "LTSpice download".

After the installation is complete,
Install the previous models as directed - .sub files in the .sub dir, and .asy files in the .sym dir.
Then download and un-zip Standard.zip to your \SwitcherCad\lib\cmp directory. Rename the original standard.dio file to standard.bak before you do this.

You will also need pot.zip (attached)
Place the .sub file in your .sub directory, and the .asy in your .sym directory.

Then download LM723 0v-30v Regulator.asc to your Switchercad directory.

I can't tell you offhand what the default directory would be, because practically nothing on my computer is standard.

You will then be able to run the simulation, and see what voltages you should be reading.
 

Attachments

Thread Starter

ChrisChemist116

Joined Mar 13, 2009
79
Thanks for assistance i did everything as noted by moving the zip files and .sym and so on for the rest of the files into the Spice directory.

I opened the file and run the simulation. Unfortunately i got the same result as it happened on my breadboard. When measuring the voltage coming from pin 3 to GND i have 10V while on Spice it tells voltage between 10 and 11.

I think it is not possible to simulate the effect of moving physically or adjusting the other potentiometers as you can do it with real components. But i can tell that it doesn't make a difference as nothing happens if P1 is adjusted.
Is that problem related with the design?:confused:

I am attaching the simulation given screenshot . I think i did followed all the directions in the right order.
 

Attachments

ifixit

Joined Nov 20, 2008
652
Hi,


I grabbed the design file that sgtwookie posted and ran a few tests. That design has some problems. Here is a list of defects and recommendations:
  1. The transformer output is 30V RMS at 1 Amp. Minus 2V(1V each for two diodes), minus the minimum droup of the ripple voltage at 1 Amp is 10V, minus 2V for series regulator transistor headroom. Therefore the maximum output at 1A is 30V -2-10-2= 16 V.
  2. Changing the filter cap to 5600uF reduces the ripple to 3V and allows an output of 23V.
  3. Using a transformer with a 38V secondary would allow the 30V output.
  4. The current source, Q1, needs 3.5 volts of headroom to work properly, plus the drop across R8 means 4.1 Volts is required above 30V for current limiting to work right whin the output is set above 26V.
  5. Change the Vcc voltage to 35V. Caution: Since the Vcc is -4.6V, this will put the 723 near its maximum operating voltage of 40V.
  6. Adjust P3 so that it is at minimum resistance, which will allow for 2A current maximum. This will ensure that current limiting is not killing your output during your first power up tests.
  7. Adjust the P1 pot so that pin 5(non inv) is approximately 0V wrt ground.
I have more, but it is late... time for bed.


Regards,
Ifixit
 
Top