EEVblog Electronics Community Forum

Electronics => Projects, Designs, and Technical Stuff => Topic started by: echen1024 on September 01, 2013, 03:51:28 pm

Title: DC PSU
Post by: echen1024 on September 01, 2013, 03:51:28 pm
I currently have a LM317 set up in the method TI has it in the datasheet, with a rectifier, and then 2 1000uf filter caps, then the regulator. I was wondering if there is anyway to, for example, use op-amps, to further reduce the ripple voltage. Any help would be appreciated, and please, put the part numbers in.

Thanks.
Title: Re: DC PSU
Post by: rdl on September 01, 2013, 04:26:16 pm
Do you have the bypass cap on the adjust pin? Do you have sufficient load current?
Title: Re: DC PSU
Post by: echen1024 on September 01, 2013, 04:29:20 pm
Measurements are done with 0.5A of load current, and a bypass cap.
Title: Re: DC PSU
Post by: mmagin on September 01, 2013, 04:50:47 pm
I am assuming you have this set up as in Figure 1 in the datasheet.
How much ripple are you seeing on the output?

What are the input and output voltages?  The datasheet specifies a minimum differential of 3 volts between the input and output, and of course, that's going to have to be true under the expected load and the lowest points in the waveform seen on the input filter cap.
Title: Re: DC PSU
Post by: echen1024 on September 01, 2013, 04:55:29 pm
I am assuming you have this set up as in Figure 1 in the datasheet.
How much ripple are you seeing on the output?

What are the input and output voltages?  The datasheet specifies a minimum differential of 3 volts between the input and output, and of course, that's going to have to be true under the expected load and the lowest points in the waveform seen on the input filter cap.

Input is around 20 volts, and output is 5. I am seeing around 400mv p-p of ripple, though that could be the scope noise as well.
Title: Re: DC PSU
Post by: mmagin on September 01, 2013, 05:10:22 pm
Input is around 20 volts, and output is 5. I am seeing around 400mv p-p of ripple, though that could be the scope noise as well.

Wow, that is a lot.  You have input and output capacitors, and a capacitor on the adjust pin?  What values (besides the 2000 uF you already mentioned on the input)?
Title: Re: DC PSU
Post by: hlavac on September 01, 2013, 05:13:09 pm
What is the shape of the ripple waveform?
And more importantly, do you have a largish heatsink on the regulator? 7.5W is enough to fry it...
Title: Re: DC PSU
Post by: echen1024 on September 01, 2013, 06:45:30 pm
What is the shape of the ripple waveform?
And more importantly, do you have a largish heatsink on the regulator? 7.5W is enough to fry it...

I have a heatsink used for the Intel 486, with a large, brushless DC fan on the back. And I have a 10uf, 0.1uf, and a 1uf capacitor.
Title: Re: DC PSU
Post by: rdl on September 01, 2013, 07:44:02 pm
It's probably time for you to post the actual schematic you built and/or a decent photo, because 400mV of ripple is way more than you should ever get out of an LM317. Maybe the chip is actually not working?
Title: Re: DC PSU
Post by: echen1024 on September 01, 2013, 11:39:40 pm
It's probably time for you to post the actual schematic you built and/or a decent photo, because 400mV of ripple is way more than you should ever get out of an LM317. Maybe the chip is actually not working?

I investigated, and I think that most of the noise is the oscilloscope probe crap.