Proper procedure for reducing ripple on DC power rail? (e.g. oscilloscope reading -> filter design)

[Mahonroy]

Hey guys,
I have a 3.3 volt DC power rail. I have an integrated circuit that can only operate correctly if its power input has less than 30mV ripple.

What is the procedure for building the corresponding filter circuitry to accommodate those requirements?
Is there a reading I can take with the oscilloscope, then build a circuit accordingly, then verify again with the oscilloscope? Is there a way to go about designing this where I don't even need to use an oscilloscope - e.g. do some math and it should guarantee the 30mV max ripple requirements?

What about additional noise that is on the line, will it take care of that as well?

How do you guys go about this, whats the process?

Here are some more specifics if they are needed for this example:

• Integrated circuit has a max current draw spikes of 40mA, pretty much anywhere from 0.1mA to 40mA. Averaging 20mA.
• Lets say the voltage regulator can handle a max output of 250mA.
• 3.3V, 30mV ripple max

 

[davenn]

I have an integrated circuit that can only operate correctly if its power input has less than 30mV ripple.

30mV is pretty tiny .... what IC ?

there are many links to info on the net, here's a couple, you can look for others

http://www.skillbank.co.uk/psu/smoothing.htm

http://www.irjes.com/Papers/vol2-issue6/Version-1/E02064249.pdf


Dave

 

[(*steve*)]

Additionally, the current shouted may be higher than you expect if you are measuring them using a multimeter.

Are you using a switch mode regulator or a linear regulator to supply the 3v3?

 

[Mahonroy]

30mV is pretty tiny .... what IC ?

there are many links to info on the net, here's a couple, you can look for others

http://www.skillbank.co.uk/psu/smoothing.htm

http://www.irjes.com/Papers/vol2-issue6/Version-1/E02064249.pdf


Dave

Thanks for the info, I will check out those pages.
Yeah I thought 30mV was pretty small too. This is for a nRF24L01+ module. I spoke to the engineers of some of the modules, and they said that the power input ripple voltage cannot exceed 30mV or else the transmission will not work.

Additionally, the current shouted may be higher than you expect if you are measuring them using a multimeter.

Are you using a switch mode regulator or a linear regulator to supply the 3v3?

When I measure with a voltmeter I get an average of 20mA, but after doing some reading the spikes can be a bit higher so I figured the 40mA would be safe?

I have 3 separate designs I'm working with. I am wondering if the process would be the same for each design?

• 12VDC power adapter into a 5V switcher (RECOM r-78e5.0-1.0). There is then a 3.3V LDO voltage regulator (MCP1700) connected to the 5V rail.
• 12VDC power adapter into a 3.3V switcher (RECOM r-78e3.3-0.5).
• LNK305 switcher generating a 4.5V rail. Then the 3.3V LDO voltage regulator (MCP1700) connected to this 4.5V rail.

 

[(*steve*)]

I recommend you place a 10uF, a 1uF, and a 0.1uF capacitor near the power pins of the nrf24l01. This is in addition to the capacitor near the 3v3 regulator

 

[Mahonroy]

I recommend you place a 10uF, a 1uF, and a 0.1uF capacitor near the power pins of the nrf24l01. This is in addition to the capacitor near the 3v3 regulator

Hi Steve, thanks for the response!
I have actually already tried this. This morning I tried splicing in an RC filter on the 5V rail, and here are the results (not much difference):

BEFORE:

Video of oscilloscope:
https://goo.gl/photos/TxpybHEWfKQrqGRp7
https://goo.gl/photos/h5W5WoXW97UWjsgS8

AFTER:

Video of oscilloscope:
https://goo.gl/photos/A2G4f1x3rvNkj4dt8
https://goo.gl/photos/BZsGnWFYK9TVxs418

What do you guys think?

 

[Mahonroy]

I took a new board without the RC modification, completely severed the circuit between C8 and R4, rigged up a 7805 and a power adapter. Here are the results:
https://goo.gl/photos/U3rpHXU7YF95NMjt5

I now replaced the power adapter with a 9 volt battery:
https://goo.gl/photos/gFQuckUUTys6Pyzo9

Wow a lot cleaner. I'm confused about the occasional big dips that show up on the scope as well.

 

[(*steve*)]

Unfortunately I cant see the images on my phone, so things are not very clear to me.

The problem could be nose from your adapter, your pre-regulator, or even your 3.3v reg.

I recommend you look at the power supply rails (under load) from each.

One possible problem is poor grounding in your smps leading to noise bring injected into the ground rail. Did you design the PCB for your regulator? Can you post images of them (here)?

Another possible problem is the use of capacitors with too high an ESR around the SMPS.

 

[Mahonroy]

Unfortunately I cant see the images on my phone, so things are not very clear to me.

The problem could be nose from your adapter, your pre-regulator, or even your 3.3v reg.

I recommend you look at the power supply rails (under load) from each.

One possible problem is poor grounding in your smps leading to noise bring injected into the ground rail. Did you design the PCB for your regulator? Can you post images of them (here)?

Another possible problem is the use of capacitors with too high an ESR around the SMPS.

Hi Steve,
I did design the SMPS, here is the schematic and layout of it:


U$2 is LNK305
C6, C7 is this part number: 475CKE450MLN
C8 is this part number: 35ZLH100MEFC6.3X11

Thanks again for your help!

 

[davenn]

Hi Steve,
I did design the SMPS, here is the schematic and layout of it:

U$2 is LNK305
C6, C7 is this part number: 475CKE450MLN
C8 is this part number: 35ZLH100MEFC6.3X11

Thanks again for your help!

well that's pretty shocking in more ways than one
holy crap .... you have absolutely NO mains isolation !!!

Personally, I would NEVER use one of those switcher chips where the final outcome voltages are available for touching
by whoever may use the device. What you are doing is a huge safety risk if any one part in the initial sections of the PSU
should fail in short circuit mode.

The rectification of the 120VAC could hardly be called full wave and as a result the ripple present is going to be horrific

I appreciate you may be wanting compactness, but that should come at the cost of safety

Dave

 

[Mahonroy]

well that's pretty shocking in more ways than one
holy crap .... you have absolutely NO mains isolation !!!

Personally, I would NEVER use one of those switcher chips where the final outcome voltages are available for touching
by whoever may use the device. What you are doing is a huge safety risk if any one part in the initial sections of the PSU
should fail in short circuit mode.

The rectification of the 120VAC could hardly be called full wave and as a result the ripple present is going to be horrific

I appreciate you may be wanting compactness, but that should come at the cost of safety

Dave

Don't worry Dave, I did my homework on this and its ok... this device is completely sealed inside an enclosure with absolutely no interface to the outside world... its completely sealed, no buttons no nothing. I also incorporated the fuse and wire-wound resistor as additional safety recommended by the datasheet and certifications.

 

[(*steve*)]

What is D4 there for?

Ant reason not to use full wave rectification?

I cannot recommend taking any measurements on this device using a mains powered oscilloscope. Are you using an isolating transformer?

 

[Mahonroy]

What is D4 there for?

Ant reason not to use full wave rectification?

I cannot recommend taking any measurements on this device using a mains powered oscilloscope. Are you using an isolating transformer?

This is actually a full wave rectification power setup. For D4, are you wondering why its in that location? Or why its in the design period? I just followed the datasheet for the lnk304/305/306 and researched all of the requirements for the parts it recommended (assuming I did it correctly haha!)

And yes, I have an isolation transformer I plug this thing into so I can scope it just fine. (more specifically, I use one of those 120V battery backup blocks that people use for their computers for when the power goes out... works really well for this).

 

[(*steve*)]

No, you have half wave rectification.

A link to the datasheet would be handy, but I'll go looking.

 

[Mahonroy]

No, you have half wave rectification.

A link to the datasheet would be handy, but I'll go looking.

Sorry Steve, here is the datasheet:
http://www.mouser.com/ds/2/328/lnk302_304-306-179954.pdf

 

[davenn]

This is actually a full wave rectification power setup.

as I said and Steve confirmed, it is only 1/2 wave rectification
How do we know this you ask ? ... look at the rectifier diodes D3 and D4, there is a current path for only 1 half of the AC cycle
and that is in on the HOT through D3 and the switcher and out on the neutral via D4. The other 1/2 cycle would come in on the Neutral
go through the circuitry and out on the HOT, but it cant because it gets blocked by D4 so is never rectified.

So as I said, you have a huge source of ripple because of the 1/2 wave rectification and I didn't seen any obvious comments in the datasheet
for ripple rejection
Now you also have a second significant source of noise, and that is the switcher chip itself and that will be represented by a high freq AC signal
on the outgoing DC rail.

These switcher chips are not designed for your style of application because of their "dirty" output. They are designed for where a rough basic
DC rail is needed and noise isn't a concern

Don't worry Dave, I did my homework on this and its ok... this device is completely sealed inside an enclosure with absolutely no interface to the outside world... its completely sealed, no buttons no nothing

I'm not convinced of that considering the things you have overlooked

really ? are you sure ? .... what about the antenna for the TX module ? is it fully encased as well ?
and what about those 6 datalines to the TX module ? where do they come from ? ie. what is feeding the data to the TX ?

Dave

 

[Mahonroy]

as I said and Steve confirmed, it is only 1/2 wave rectification
How do we know this you ask ? ... look at the rectifier diodes D3 and D4, there is a current path for only 1 half of the AC cycle
and that is in on the HOT through D3 and the switcher and out on the neutral via D4. The other 1/2 cycle would come in on the Neutral
go through the circuitry and out on the HOT, but it cant because it gets blocked by D4 so is never rectified.

So as I said, you have a huge source of ripple because of the 1/2 wave rectification and I didn't seen any obvious comments in the datasheet
for ripple rejection
Now you also have a second significant source of noise, and that is the switcher chip itself and that will be represented by a high freq AC signal
on the outgoing DC rail.

These switcher chips are not designed for your style of application because of their "dirty" output. They are designed for where a rough basic
DC rail is needed and noise isn't a concern



I'm not convinced of that considering the things you have overlooked

really ? are you sure ? .... what about the antenna for the TX module ? is it fully encased as well ?
and what about those 6 datalines to the TX module ? where do they come from ? ie. what is feeding the data to the TX ?

Dave

Yeah it does look like 1/2 wave now that I look at it. I wonder why the datasheet doesn't mention this stuff? The switcher chip works by skipping the pulses, and this is the only configuration in the datasheet, so wouldn't providing it double the waves mess something up? After some googling it looks like most people use the chip completely differently than whats in the datasheet... pretty confusing.

There is a microcontroller that communicates with the RF module.

Can't this dirty output be filtered down to a usable form?

 

[davenn]

There is a microcontroller that communicates with the RF module.

and I am assuming the microcontroller and whatever it controls is outside the box ?
and you didn't answer my comment about the antenna ?

both these are places where some one could be exposed to mains voltage in fault conditions

Can't this dirty output be filtered down to a usable form?

not likely to the extent you require. .... as I said, this particular type of switcher supply is designed for where a cheap and nasty
DC supply is required that is TOTALLY isolated from the outside world

You really, REALLY need to consider using a more standard PSU


Dave

 

[Mahonroy]

and I am assuming the microcontroller and whatever it controls is outside the box ?
and you didn't answer my comment about the antenna ?

both these are places where some one could be exposed to mains voltage in fault conditions



not likely to the extent you require. .... as I said, this particular type of switcher supply is designed for where a cheap and nasty
DC supply is required that is TOTALLY isolated from the outside world

You really, REALLY need to consider using a more standard PSU


Dave

Nope, everything is inside the box. Including the antenna (its a chip antenna on the PCB). It controls a wall outlet. So you plug this device into the wall, you then plug whatever you want into this device, and then you can control it via RF. There are no switches or antennas or anything, just a input plug and output plug.

Can you give me some recommendations on some more standard PSUs? When I started researching this it looked to be a popular one so I chose it.

By the way, it turns out my oscilloscope has like 30mV worth of noise on it for some reason...
I disconnected the AC stuff from the board (dremel tool), and rigged it up to a 7805 linear voltage regulator and 9v battery. I still have a wave that is 30-50mV thick. I then disconnected the entire circuit and just left with 7805 linear regulator and battery and it still looked the same...

 

[(*steve*)]

Are you using you oscilloscope properly?

What does the trace look like if you connect the ground clip to the probe tip?

What if you connect it directly across a filter cap on your 7805? (you are using filter caps, right?)