Voltage Regulators - Sensitivity?

[westers151]

Hi

I require a voltage regulator which has 24V at the output, however my input is 24V AC/DC * 1.4 = 33.6V DC.

The regulators I'm looking at (LM 7824, etc), have a Vin of 33V for a Vout of 24V.

With my rectified voltage being 33.6V, how critical is the extra 0.6V to the operation of the LM 7824?

Thanks

 

[Bill_P]

The Fairchild LM7824 datasheet indicates that certain guaranteed specifications are measured at 33V input. Operationally the regulator works well with inputs ranging from 27V to 38V as indicated in the line regulation specification.

 

[jan.didden]

Hi

I require a voltage regulator which has 24V at the output, however my input is 24V AC/DC * 1.4 = 33.6V DC.

The regulators I'm looking at (LM 7824, etc), have a Vin of 33V for a Vout of 24V.

With my rectified voltage being 33.6V, how critical is the extra 0.6V to the operation of the LM 7824?

Thanks

You were looking at the max input voltage. What you need is the so called drop-out voltage. That is the min input-output differential required for correct operation.
Probably about 2V, so you need to be sure that the min input is at least 26V.
The min input occurs at the valley of the ripple so need to be sure that the input ripple is not too high.

The 33V max input can be an issue when you switch on the supply and the output cap is uncharged, meaning the full input voltage is across the reg.
But from 24AC you won't get 33.6VDC because there is at least one diode drop getting you back to 33V, and with a bridge rectifier two diode drops getting you down to nominally 32.4. OTOH, a xformer loaded less than nominal can output up to 10% more than the nominal 24VAC...
Anyway, I believe that most 24V regs have a max Vin of 40V like the Fairchild LM7824, so don't know where your 33V comes from....

jan

 

[DUG]

You did not mention how much current you were going to load this with.

However.

According to the data sheet I looked at ( http://www.fairchildsemi.com/ds/LM/LM7805.pdf ) The dropout of a LM7824 at 1A is 2V. So you can go down to 26V before the output will be affected. ( at 1A)

So, your 24 x 1.414 - 1.2 = 32.7Vp

At 1 amp the filter cap needs to be a minimum of 1000uF (roughly calculated 975uF)

Keeping close to this value will minimize regulator power dissipation. The trade-off will be regulator supply ripple. Which, at 54dB rejection( data sheet at .5A), will show up at less than 20mV at the output. (Add caps on the output to reduce this)

Increasing the filter cap size will:

Increase regulator power dissipation
Increase rectifier peak currents
Decrease regulator supply ripple feedthrough.

Very rough calculation for minimum power dissipation for this setup:

0.5A 500uF
0.75A 750uF
1A 1000uF
1.5A 1600uF

If you have adequate heat sinking then add 50% to the above capacitor values.

That is what I would do.

 

[RJM1]

You could always add a resistor in series with the secondary of the power transformer to reduce the voltage (resistor value to be determined by average current draw.). This would also reduce the peak currents in the rectifier diodes, ripple, hf ripple and RF emissions.

 

[DUG]

And regulator power dissipation.