LM723 DESIGN TIP

Thread Starter

Exjay

Joined Nov 19, 2015
196
I want to design a lab power supply with voltage ranges between 0-15v and 1.5A. So I came to the conclusion of using LM723. I read the IC datasheet but I don't know how to go about it. Please I need explanation of the process. I don't want any circuit. I want to develop from scratch. Please am an amateur hobbyist.
 

Papabravo

Joined Feb 24, 2006
22,082
The default assumption is that the datasheet is correct. If you can prove otherwise, I'm sure the manufacturer would like to hear about it. If you want to develop something from scratch, stop talking about it and just do it. It is the only way. There are no shortcuts, no easy peasy do this 1-2-3. Put out a schematic and ask for comments.
 

Analog Ground

Joined Apr 24, 2019
460
Isn't the 723 over 40 years old? As in the Fairchild uA723? I cannot imagine it is recommended for new designs. Back when, it was painful! Should the TS reconsider?
 

crutschow

Joined Mar 14, 2008
38,503
You might look at some of these designs.

But for a much simpler, rugged design I would go with a single chip design such as the LT3081.
It's not a cheap device, but will likely be cheaper than all the parts for an LM723 circuit.
It can be adjusted to a 0V output, and has an adjustable current limit, as well as outputs for the device temperature and output current indication (avoiding the need for a current shunt-resistor).
I discuss such a design here.
 

Thread Starter

Exjay

Joined Nov 19, 2015
196
Thank you all. I have started the prototype on breadboard. I have problem comprehending something in the datasheet. If I want to build a variable voltage I used the table with resistor value and the variable output column. Am I in line? Please I need guidance
 

crutschow

Joined Mar 14, 2008
38,503
Thank you all. I have started the prototype on breadboard. I have problem comprehending something in the datasheet. If I want to build a variable voltage I used the table with resistor value and the variable output column. Am I in line? Please I need guidance
What data sheet?
 

JohnInTX

Joined Jun 26, 2012
4,787
Isn't the 723 over 40 years old? As in the Fairchild uA723? I cannot imagine it is recommended for new designs. Back when, it was painful! Should the TS reconsider?
Believe it or not, it's still current on ti.com.

@Exjay You should always post the datasheet or at least a link if you want help on a particular device. You should also post a schematic of what you are proposing to do. It's asking a little much for forum members to do designs for free on request.
 

Attachments

schmitt trigger

Joined Jul 12, 2010
2,087
Exjay:
What Crutschow is most likely saying, is that when making a question, it is considered good forum etiquette to attach all documents pertinent to the discussion. Among them, the datasheet.

And I know, I know.............. datasheets are readily available online.
But if you are asking the question, avoid the extra work and time to anyone who may help you out.
 

Thread Starter

Exjay

Joined Nov 19, 2015
196
Thank you John. I'm completely lost. My question is how to generate a circuit diagram for 0-15v power supply using the LM723 and its datasheet
 

Thread Starter

Exjay

Joined Nov 19, 2015
196
Table 1 of that datasheet is the problem. I want a maximum 15v. What diagram will I use in the datasheet? I will be very grateful if I fully implement it with your helps
 

schmitt trigger

Joined Jul 12, 2010
2,087
Table 1 has a column which tells you which circuit configuration to use: APPLICABLE FIGURES SEE NOTE 5

And Note 5 says:
NOTES: 5. The R1/R2 divider can be across either VO or V(ref). If the divider is across V(ref), use the figure numbers without parentheses. If the divider is across VO, use the figure numbers in parentheses.

The important thing from this note, is that at about 7 volts there is an operational transition, and the resistor divider has to move from Vref to Vout. In plain English, you cannot do a seamless, continuously variable regulator from 2 to 15 volts.
Additionally, 2 volts is the minimum output voltage without a negative bias.
 

crutschow

Joined Mar 14, 2008
38,503
at about 7 volts there is an operational transition, and the resistor divider has to move from Vref to Vout. In plain English, you cannot do a seamless, continuously variable regulator from 2 to 15 volts.
Although the LM723 data sheet doesn't explicitly show how to make an adjustable supply continuously variable from its minimum regulated voltage (2V) to more than 7V, that doesn't mean you can't.
You can extend the range by essentially combining the two circuits.

Below is the LTspice simulation of an LM723 circuit that can go from 2V to any high voltage within its limits:
The circuit values shown give an output range of 2V to >15V (blue trace) for pot U2 adjustment of 0 to 100% (0 to 1 horizontal scale).

The 723 regulates by maintaining the voltage on the Inv input to be equal to the Ninv input voltage (which look like the inputs to an internal op amp).
The R1 and R2 divider gives a voltage of 2V at the Ninv input from the 7V reference, Vref.
U2 and R3 act as a voltage divider to determine the percentage of the output voltage that appears at the Inv terminal.
So the output is adjusted by the 723 until the output voltage generates 2V at this Inv terminal.

Of course to get to 1.5A you will need to add a buffer power transistor as the data sheet shows.

1774297330832.png


3/22/26 Edit: Since this posting I have developed a simple LM723 circuit that can go down to zero volts output (below);

It uses the 7V Ref voltage to buck the 2V Inv feedback voltage at the adjustment pot output to allow the output to go to zero when the pot wiper is at the top (0%) position.
The PNP-NPN buffer stage allows a current output of at least 3A (here limited to 3a by Rcl).

1774297049000.png
 
Last edited:
Top