Author Topic: How to reduce the big ripple at the input of 7805?  (Read 5659 times)

0 Members and 1 Guest are viewing this topic.

Offline nadona

  • Regular Contributor
  • *
  • Posts: 205
  • Country: us
How to reduce the big ripple at the input of 7805?
« on: January 15, 2015, 01:37:58 am »
I have a Solartron 7061 which has big ripple at the input of 7805. Even though the original 10000uF 16V cap was good, I changed with a new 12000 50V Nichion cap to see if the ripple change. The noise was reduced to about 25mV at Vcc pin of GPIB but ripple size is same.

How do I reduce the ripple :-// Or it doesn't matter as long as Vcc is clean and accurate to +5V?

First picture shows the size and frequency of the ripple.
The red line is where big ripple is measured in the second picture.

Any suggestion will be appreciated.

[Edit] Corrected the capacitors size & noise at the GPIB Vcc pin
« Last Edit: January 15, 2015, 04:42:29 am by nadona »
Ha-ha-ha. That's good, too!
 

Offline c4757p

  • Super Contributor
  • ***
  • Posts: 7805
  • Country: us
  • adieu
Re: How to reduce the big ripple at the input of 7805?
« Reply #1 on: January 15, 2015, 01:40:13 am »
Less than a volt, at the output of a rectifier? That's how rectifiers work.
No longer active here - try the IRC channel if you just can't be without me :)
 

Offline Fsck

  • Super Contributor
  • ***
  • Posts: 1157
  • Country: ca
  • sleep deprived
Re: How to reduce the big ripple at the input of 7805?
« Reply #2 on: January 15, 2015, 01:40:32 am »
I'm confused about why you care about ripple before the regulator
"This is a one line proof...if we start sufficiently far to the left."
 

Offline nadona

  • Regular Contributor
  • *
  • Posts: 205
  • Country: us
Re: How to reduce the big ripple at the input of 7805?
« Reply #3 on: January 15, 2015, 01:55:47 am »
c4757p:
Thank you.

Fsck:
I never saw this big ripple before and some of logic ic pins have unusually big noise. And  I changed the cap 50% bigger so expected smaller ripple.

I am trying to figure out if this has anything to do with why the GPIB controller's interrupt pin is active LOW all the time from the moment the DMM boot-up and MPU can't access RESET vector.

Thank you for asking.
« Last Edit: January 15, 2015, 02:03:52 am by nadona »
Ha-ha-ha. That's good, too!
 

Offline wraper

  • Supporter
  • ****
  • Posts: 10208
  • Country: lv
Re: How to reduce the big ripple at the input of 7805?
« Reply #4 on: January 15, 2015, 02:07:22 am »
I am trying to figure out if this has anything to do with why the GPIB controller's interrupt pin is active LOW all the time from the moment the DMM boot-up and MPU can't access RESET vector.
No, at least not ripple itself at that point.
 

Offline nadona

  • Regular Contributor
  • *
  • Posts: 205
  • Country: us
Re: How to reduce the big ripple at the input of 7805?
« Reply #5 on: January 15, 2015, 02:20:43 am »
I am trying to figure out if this has anything to do with why the GPIB controller's interrupt pin is active LOW all the time from the moment the DMM boot-up and MPU can't access RESET vector.
No, at least not ripple itself at that point.

Thank you ;D
Ha-ha-ha. That's good, too!
 

Offline digsys

  • Supporter
  • ****
  • Posts: 2037
  • Country: au
    • DIGSYS
Re: How to reduce the big ripple at the input of 7805?
« Reply #6 on: January 15, 2015, 02:21:14 am »
Something is a bit fishy to me ...
There should be negligible ripple on the Input side with NO load. Can you check that?
You went from 1,000 to 15,000uF and saw NO reduction ??? What was the current at that time? Can you estimate it, if you don't know exactly?
Do you have 0.1uF ceramic caps on the Input and Output of the Reulator AND/OR ~4.7-10uF Tantalums or similar ESR?
Is the MPU circuit any distance from the Regulator? Do you have a 0.1uF cap on the MPU PS pins? Any other 0.1uF caps?
It does seem excessive to me at least.
Hello <tap> <tap> .. is this thing on?
 

Offline nadona

  • Regular Contributor
  • *
  • Posts: 205
  • Country: us
Re: How to reduce the big ripple at the input of 7805?
« Reply #7 on: January 15, 2015, 03:00:03 am »
Something is a bit fishy to me ...
There should be negligible ripple on the Input side with NO load. Can you check that?
You went from 1,000 to 15,000uF and saw NO reduction ??? What was the current at that time? Can you estimate it, if you don't know exactly?
Do you have 0.1uF ceramic caps on the Input and Output of the Reulator AND/OR ~4.7-10uF Tantalums or similar ESR?
Is the MPU circuit any distance from the Regulator? Do you have a 0.1uF cap on the MPU PS pins? Any other 0.1uF caps?
It does seem excessive to me at least.

digsys:
Sorry, the capacitor was wrong. I corrected the original cap size to 10000uF 16V. Good catch. The another reason changing the cap was because the old one has DF of 2.3.

I will check again about the small caps.
Ha-ha-ha. That's good, too!
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 9556
  • Country: us
Re: How to reduce the big ripple at the input of 7805?
« Reply #8 on: January 15, 2015, 03:50:24 am »
OK, you can do a calculation to estimate the load current on the filter capacitor.

The equation to use is:

    I = C * dV/dt

According to your post the capacitor is 12000 µF, and from the scope trace dV/dt is 0.68 V / 5 ms = 136 V/s. Plugging in the numbers gives:

    I = 12000 x 10-6 x 136 = 1.6 A

Assuming I haven't made a mistake, that seems like a really high current for a poor 7805 to provide.

If I'm right, you probably have a faulty component down circuit from the regulator causing an excessive current draw, and that is why you are seeing such a large ripple voltage.
I'm not an EE--what am I doing here?
 

Offline nadona

  • Regular Contributor
  • *
  • Posts: 205
  • Country: us
Re: How to reduce the big ripple at the input of 7805?
« Reply #9 on: January 15, 2015, 04:28:14 am »
OK, you can do a calculation to estimate the load current on the filter capacitor.

The equation to use is:

    I = C * dV/dt

According to your post the capacitor is 12000 µF, and from the scope trace dV/dt is 0.68 V / 5 ms = 136 V/s. Plugging in the numbers gives:

    I = 12000 x 10-6 x 136 = 1.6 A

Assuming I haven't made a mistake, that seems like a really high current for a poor 7805 to provide.

If I'm right, you probably have a faulty component down circuit from the regulator causing an excessive current draw, and that is why you are seeing such a large ripple voltage.

IanB:
As you can see in the first post's picture of power supply schematics, there are 4 7805s all on the earthy logic board and one more cap of 3300uF feeding 4 7805s

GPIB's Vcc out of IC504 is clean but MPU's Vcc out of IC502, I just found out, is very noisy. C510 at the output of IC502 is 23uF and I put 0.1uF ceramic cap under the board across the pin1(Vss) and pin7(Vcc) but noise came down just a little bit. My LeCroy crashed...coffee break now ;D  I will try different size of cap underneath 68B09 when LeCroy cool down.

Thank you for the calculation.
« Last Edit: January 15, 2015, 05:12:20 am by nadona »
Ha-ha-ha. That's good, too!
 

Offline Rupunzell

  • Frequent Contributor
  • **
  • Posts: 349
Re: How to reduce the big ripple at the input of 7805?
« Reply #10 on: January 19, 2015, 04:55:18 pm »
There are versions-brands (yes, three terminal regulators are not created equal)  of 7805 and other three terminal regulators that dump a lot of noise current back into the ground via it's ground pin. Have a look at how the actual ground connections are made into the regulator's ground pin and the rest of the system. If there is path that would allow the ground current noise to appear from the regulator, that could be part of this problem.

A larger filter cap or bypassing with more caps are not going to cure this problem and bypassing can actually make the problem worst as caps tend to be voltage to current converters. Where this current returns does make a difference.

Another solution would be to replace the 7805 with a LM340 series three terminal regulator as they are less prone to doing stuff like this and overall an improved design.



Bernice
 

Offline Richard Crowley

  • Super Contributor
  • ***
  • Posts: 4310
  • Country: us
  • KE7GKP
Re: How to reduce the big ripple at the input of 7805?
« Reply #11 on: January 19, 2015, 05:07:38 pm »
The ripple you are seeing on the RAW output of the rectifier and bulk filter cap(s) is completely normal, perhaps even better than average.
Furthermore, such small ripple on the INPUT side of a 3-terminal regulator should have NO significant effect on its operation or the quality of its output.
Unless the average DC voltage violates the minimum input voltage of the regulator.

If you have noisy power bus problems elsewhere in the circuit, the cause (and remedy) will be found much farther DOWN-stream from there.
You are looking in the wrong place.  You have confirmed that the power supply is completely normal.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf