R-value in my simple PSU

tobias_svensk · 2nd August 2005, 09:31 PM

Hello

I have made a PCB for a simple 317/337 PSU
http://www.ettnet.se/~tobias/diy/opabuf/schema_psu.jpg

On the PCB there's only a RC and not RCRC, one of this boards is gonna drive a one-channel OPA627/BUF634 combo @ +/-17V

My question is: Is 15ohms a bit to high in the "RC"? should it be like 5ohms or something instead?

Trannies would prpbably be 2x18VAC or 2x15VAC if the voltagedrop isn't to high.

Cheers

Loial · 3rd August 2005, 08:25 AM

Hi
The size of the resistors set the LP-filter value for the RC combination, to knock out 50Hz, the lower values of R - the higher your low pass frequency cut-out gets.
At 5 ohms, your -3 dB is at approx. 26 Hz, and at 15 ohms, it is 9 Hz. Quite a difference.
Then of course, it is interesting to know how much current your the OPA627/BUF634 will draw, to minimize the losses in these resistors.

tobias_svensk · 3rd August 2005, 10:06 AM

Hello Loial

Well the combo can put out about 250mA i guess but as it's gonna be used as a preamp for a chipamp the currentdraw would not be big.

Hm, so a 2.5 ohms could work out fine then? but then it would be aiming for the 50Hz ripple? Thought I read that a higher value in an CRC would make a better overall PSRR (or what it's called)

Thanks

Loial · 3rd August 2005, 04:45 PM

I would say you should use a resistor as large as possible, to cut of the 50 Hz from the voltage.
Though a larger resistor also means larger voltage drop...

If it is used as a preamp there should'n be a problem using the 15 ohm resistors.

AndrewT · 6th August 2005, 08:18 AM

Hi,
since you are deleting the second RG, then I suggest that you add the two inline resistors together to give 30R (or use 33R).
The time constant is only 18mS (15R) and 40(33R) so there is some leaway to increase R further. Maybe try 47R.
At 10mA draw it loses half a volt. One opamp could draw 20 to 50mA into low loads at peak and volts lost before reg would be 2.4Volts at peak current (15Vpk into 400R).

cerrem · 6th August 2005, 10:21 AM

What you have is two consecutive first order RC filters..
THe result is a DOUBLE POLE filter..with a -40dB rolloff from the POLE.... In your case the pole is at 8.8Hz....
You are trying to filter 100Hz not 50Hz...since you have a full-wave bridge.....
You already know that you will be -40dB down at 88Hz..so figure it at 100Hz....
Don't sweat over this....
The input cap is figured based on the minimum DC input voltage to the linear regulator...since you have a PEAK of the ripple is refered to MAX voltage..the cap size you choose will change the Vmin of the ripple.... Also too big of a cap will shorten the conduction time/angle on the diodes and increases current peaks... The EMI and Power factor need to be assessed...

Chris

peranders · 6th August 2005, 11:49 AM

Quote:

Originally posted by tobias_svensk
Hello

I have made a PCB for a simple 317/337 PSU
http://www.ettnet.se/~tobias/diy/opabuf/schema_psu.jpg

On the PCB there's only a RC and not RCRC, one of this boards is gonna drive a one-channel OPA627/BUF634 combo @ +/-17V

My question is: Is 15ohms a bit to high in the "RC"? should it be like 5ohms or something instead?

Trannies would prpbably be 2x18VAC or 2x15VAC if the voltagedrop isn't to high.

Cheers

The resistor value isn't particular sensitive. The highest value is determined how much voltage you can afford to loose. 1-10 ohms is pretty normal to achieve some filtering. The best way to figure out the optimal value is to test.

Eva · 6th August 2005, 11:53 AM

These regulators and op-amps already show pretty high PSRR figures at low frequencies so I think there is not much left to improve. Humm due to rail ripple will be probably below noise floor anyway.

Poles may be more useful when placed around 150..300Hz, so that line harmonics and EMI get attenuated without sacrifying too much efficiency. Such poles are obtained with R values in the range 2r2..4r7 for 1200uF capacitors.

Splitting resistance between +Vcc/-Vcc and ground may be useful for EMI isolation.

jackinnj · 6th August 2005, 01:22 PM

you might find that it is less expensive, (it will also take up less real estate) -- to use a tracking pre-regulator -- National will show you how in the LM317 product folder -- a tracking pre-regulator was used in the later versions of the Super-Regulator.

tobias_svensk · 7th August 2005, 02:38 PM

Thanks for all your answers! very useful.

Perhaps I just go for a 5 ohms, it's not like it's a super psu anyway

Have another schematic I was wondering about, would this be a better psu?

Thanks

2nd August 2005, 09:31 PM	#1
tobias_svensk diyAudio Member Join Date: Jan 2004 Location: Hudiksvall	R-value in my simple PSU Hello I have made a PCB for a simple 317/337 PSU http://www.ettnet.se/~tobias/diy/opabuf/schema_psu.jpg On the PCB there's only a RC and not RCRC, one of this boards is gonna drive a one-channel OPA627/BUF634 combo @ +/-17V My question is: Is 15ohms a bit to high in the "RC"? should it be like 5ohms or something instead? Trannies would prpbably be 2x18VAC or 2x15VAC if the voltagedrop isn't to high. Cheers

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Simple Simple SE questions	nodiak	Tubelab	57	2nd February 2010 11:02 PM
Simple, SIMPLE Digital volume control	-_nando-_	Digital Source	5	21st December 2006 01:20 PM
A simple question from a simple man.	inteificio	Solid State	5	17th October 2005 03:00 PM
Simple (or not so simple) decision	wxn	Chip Amps	4	5th November 2004 09:55 AM
Simple answer required for a simple question...	rabbitz	Multi-Way	4	31st August 2003 10:05 AM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

3rd August 2005, 08:25 AM	#2
Loial diyAudio Member Join Date: Nov 2004 Location: Gävle	Hi The size of the resistors set the LP-filter value for the RC combination, to knock out 50Hz, the lower values of R - the higher your low pass frequency cut-out gets. At 5 ohms, your -3 dB is at approx. 26 Hz, and at 15 ohms, it is 9 Hz. Quite a difference. Then of course, it is interesting to know how much current your the OPA627/BUF634 will draw, to minimize the losses in these resistors.

3rd August 2005, 10:06 AM	#3
tobias_svensk diyAudio Member Join Date: Jan 2004 Location: Hudiksvall	Hello Loial Well the combo can put out about 250mA i guess but as it's gonna be used as a preamp for a chipamp the currentdraw would not be big. Hm, so a 2.5 ohms could work out fine then? but then it would be aiming for the 50Hz ripple? Thought I read that a higher value in an CRC would make a better overall PSRR (or what it's called) Thanks

3rd August 2005, 04:45 PM	#4
Loial diyAudio Member Join Date: Nov 2004 Location: Gävle	I would say you should use a resistor as large as possible, to cut of the 50 Hz from the voltage. Though a larger resistor also means larger voltage drop... If it is used as a preamp there should'n be a problem using the 15 ohm resistors.

6th August 2005, 08:18 AM	#5
AndrewT diyAudio Member Join Date: Jul 2004 Location: Scottish Borders	Hi, since you are deleting the second RG, then I suggest that you add the two inline resistors together to give 30R (or use 33R). The time constant is only 18mS (15R) and 40(33R) so there is some leaway to increase R further. Maybe try 47R. At 10mA draw it loses half a volt. One opamp could draw 20 to 50mA into low loads at peak and volts lost before reg would be 2.4Volts at peak current (15Vpk into 400R).

6th August 2005, 10:21 AM	#6
cerrem diyAudio Member Join Date: Jun 2005 Location: San Diego, CA	What you have is two consecutive first order RC filters.. THe result is a DOUBLE POLE filter..with a -40dB rolloff from the POLE.... In your case the pole is at 8.8Hz.... You are trying to filter 100Hz not 50Hz...since you have a full-wave bridge..... You already know that you will be -40dB down at 88Hz..so figure it at 100Hz.... Don't sweat over this.... The input cap is figured based on the minimum DC input voltage to the linear regulator...since you have a PEAK of the ripple is refered to MAX voltage..the cap size you choose will change the Vmin of the ripple.... Also too big of a cap will shorten the conduction time/angle on the diodes and increases current peaks... The EMI and Power factor need to be assessed... Chris

6th August 2005, 11:53 AM	#8
Eva diyAudio Member Join Date: Oct 2003 Location: Near the sea	These regulators and op-amps already show pretty high PSRR figures at low frequencies so I think there is not much left to improve. Humm due to rail ripple will be probably below noise floor anyway. Poles may be more useful when placed around 150..300Hz, so that line harmonics and EMI get attenuated without sacrifying too much efficiency. Such poles are obtained with R values in the range 2r2..4r7 for 1200uF capacitors. Splitting resistance between +Vcc/-Vcc and ground may be useful for EMI isolation.

6th August 2005, 01:22 PM	#9
jackinnj diyAudio Member Join Date: Apr 2002 Location: Llanddewi Brefi, NJ	you might find that it is less expensive, (it will also take up less real estate) -- to use a tracking pre-regulator -- National will show you how in the LM317 product folder -- a tracking pre-regulator was used in the later versions of the Super-Regulator.

7th August 2005, 02:38 PM	#10
tobias_svensk diyAudio Member Join Date: Jan 2004 Location: Hudiksvall	Thanks for all your answers! very useful. Perhaps I just go for a 5 ohms, it's not like it's a super psu anyway Have another schematic I was wondering about, would this be a better psu? Thanks