Sulzer regulator

[starushz]

Hi, Im currently deciding what regulator design to use.

1. Jung super regulator (expensive)
2. Sulzer regualtor (cheaper but doesnt have protection cicruit)

The question i would like to ask is, is it possible to add a current limiting circuit to the sulzer (1 transistor and a resistor) without adding additional noise to the circuit?
Or is there any other designs that i can consider?

Thanks

 

[peranders]

A Jung Super Regulator isn't particulary more expensive than a Sulzer dito if you use the same parts. The basic difference is that the opamp of the Super Regulator gets it's power from the regulated side. The output stage is also a bit different to ensure the startup.

To add a current limiting is pretty easy, just add a current shunt and a transistor.

 

[starushz]

in terms of no. of component the jung has more. Harder to build hte circuit as Im using a veroboard. Anyway is the seluzar better with analog or digital supply?

 

[peranders]

Hard to say about you last question but the pcb is very important if you want to have high performance.

Maybe LM317/337 is sufficient for your needs?

 

[starushz]

Dont think i will go the LM317/337 route...since im doing a whole circuit might as well do it better. My cd player currently has 78xx/79xx. For the sulzer how important is the layout as compared to the Jung?

 

[peranders]

It's a bit less important since it's a bit slower (can't be made as fast as a Super Regulator). The most important thing is to plan how the ground should be done. Draw currents only where the voltage drops don't interfere with the rest of the circuit.

 

[jackinnj]

Dont think i will go the LM317/337 route...since im doing a whole circuit might as well do it better. My cd player currently has 78xx/79xx. For the sulzer how important is the layout as compared to the Jung?

Layout is going to affect the impedance of the Jung regulator pretty dramatically -- at least according to Walt himself -- for the lowest impedance version the ground returns were all drawn in exactly the way the authors intended.

In the last of the series of articles on the super-reg Jung used an LM317 as a constant current source, as a current regulator.

finally, in the last series of the article they did compensate the opamp, surrendering to the fact that the circuit was capable of oscillation.

you can search the archives of EDN Magazine under Walt Jung to see his 1997 article.

 

[peranders]

Jack, "dramatically", it depends from where you start. If the gentleman here tries to use an AD825 or similar I'm pretty sure he gets into trouble but a slower opamp will probably work _if_ the layout of the Vero board is as good as possible.

I must also correct you a bit. The LM317 was used as a pre-regulator creating a constant voltage over the pass transistor.

 

[starushz]

Regarding the sulzer, what is the max current draw?
I was intending to use a LM317 to preregu the voltage. By putting a fuse before the LM317, it would also limit the current ( 1A in my case). Is that correct?
Also is there any other transistor to replace the 2n3053?
How about the BC637? for negative voltage the BC638?

 

[peranders]

Max current is determined by max base current you can feed the pass transistor with, the power dissipation and the cooling and in the end max ratings for the transistors.

 

[starushz]

Sorry to ask this but fo the Sulzer circuit, how do you adjsut the output current? I see in Tangent's webpage that Vout=Vref(R2/R1+1).
In this case the Vref he is refering to is the breakdown voltage of the zener diode or what?
Thanks

 

[peranders]

Output current? You mean voltage?

From the zener you'll get a stable voltage, 5 V, 6.8 V or whatever, and then it is amplified by the opamp and the feedback network.

The zener creates a reference voltage but you can also use some other (better) circuit to get a stable voltage.

 

[starushz]

Thanks for the reply but what i meant was how to adjsut the output voltage?
looking at the circuit i think its by using the resistors at the feedback loop of the opamp

The schematic that im refering to is
http://tangentsoft.net/elec/opamp-linreg.html
(The sulzer ckt is inside)

If using the formula
Vout=Vref(R2/R1+1) (Vref is the zener voltage??)
In order to get a 5V regulator, I msut chose another zener diode and adjust the ratio of the resistor R1 and R2. Am I correct?

Thanks

 

[peranders]

You can do it in two ways:

1 Change the reference voltage, easy if you use an adjustable one.

2 Change the feedback (R1 and R2), the voltage divider from the output into the "-" input of the opamp.

 

[starushz]

Hmm..see how...will decide on cost. Anyway i've decided to use 2 transformer. One for digital and the other for analog. The orginal tranny is a toriodal (using it for analog). Should i get another toriodal or maybe get an encapsualted tranny for the digial transformer?

Thanks

 

[peranders]

Let your wallet decide here. Plain transformers will do.

 

[jan.didden]

Thanks for the reply but what i meant was how to adjsut the output voltage?
looking at the circuit i think its by using the resistors at the feedback loop of the opamp

The schematic that im refering to is
http://tangentsoft.net/elec/opamp-linreg.html
(The sulzer ckt is inside)

If using the formula
Vout=Vref(R2/R1+1) (Vref is the zener voltage??)
In order to get a 5V regulator, I msut chose another zener diode and adjust the ratio of the resistor R1 and R2. Am I correct?

Thanks

The formula is correct I think. The easiest way to visualise it is to assume that the two inputs of the opamp have the same voltage. One input (non-inverting) has Vref, the other has a fraction of the output determined by the resistive divider. That's how you get to the formula.

From the formula it is clear you can change Vout in two ways:
- either change Vref (which means Vout on the other side of the equation must change proportionally to keep both sides of the equation equal);
- or change the ratio R2 to R1 (which means Vout must change to keep the right side equal to Vref.

BTW, the description in the link you posted has some serious errors. The description of the purpose of C1 is wrong.

Jan Didden

 

[starushz]

hi can you elaborate what flaws are in the webpage that i've linked?

 

[peranders]

BTW, the description in the link you posted has some serious errors. The description of the purpose of C1 is wrong.

... or not quite the whole truth.

The large C1 cap rolls off the gain of the regulator starting at a low frequency so that high frequency noise isn't amplified by the error amp.

The main purpose is to get as much feedback as possible to reduce output errors => stable output voltage regardless of load, also reduce the noise.

 

[starushz]

What value resistors should i use? Too small and it might draw too much current. Too high and it would introduce more noise.