Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Capacitor across isolated dc/dc power pins

Status
Not open for further replies.

drkidd22

Member
Why would there be capacitors across an isolated dc/dc converter input and output terminals as shown?
 

Attachments

  • Capture.PNG
    Capture.PNG
    13.6 KB · Views: 1,702
Hy DK,

We would really need to know a lot more about you application to give a view on the capacitor's function. What component is the rectangle schematic symbol? What equipment does this portion of the schematic belong too?

spec
 
Normally you don't want capacitance between the input and output of an isolated supply, so I don't understand what those capacitors are there for. :confused:
 
Switching power supplies make noise. These capacitors help short this noise out.
This is the simple answer. The complex answer involves the inter winding capacitors in the isolation transformer.
capture-png.101777
 
Whatever the reason for the caps they pretty well defeat the usual reasons for having an isolated supply.
 
Whatever the reason for the caps they pretty well defeat the usual reasons for having an isolated supply.
No,
There is no DC path from input to output. A inductor would defeat isolation.
------------edited--------------
All AC to DC swithing power supplies have a capacitor from primary to secondary. 1000pF 1kv, power line rated. You have to have this to get noise rating. See C4 below.
upload_2016-10-25_20-19-41.png
 
Last edited:
No,
There is no DC path from input to output. A inductor would defeat isolation.
------------edited--------------
All AC to DC swithing power supplies have a capacitor from primary to secondary. 1000pF 1kv, power line rated. You have to have this to get noise rating.
I was referring to DC-DC converters which the OP's device appears to be, not line-powered AC-DC supplies.
Generally with those you are concerned about AC ground loops as well as DC loops and the capacitors gives you an AC loop.
 
All AC to DC swithing power supplies have a capacitor from primary to secondary. 1000pF 1kv, power line rated. You have to have this to get noise rating. See C4 below.
Not necesasry all switching suplies, but most of them. It is possible to make small power converters without it and still pass EMC regulations, but it is complicated and mostly trial and error to get the supply right.
 
I was referring to DC-DC converters
I was to but compared to AC to DC because many people seen the filter caps there.
The "off the shelf DC-DC bricks" that you can get have the caps inside.
It is possible to make small power converters
You will be hard pressed to find one. Even the power supplies for high end test equipment has these capacitors. I design transformeres that have low primary to secondary capacitance and noise cancelling windings and I still add the caps.
capacitors gives you an AC loop
Yes, There is noise from the transformer and I am trying to return the noise back to the source and not let it get into your loop. With out the cap the sharp edges on the MOSFET/Diodes will push through the 100s or 1000s of pF in the transformer and the noise will pound your isolated supply.

The two capacitors I added in red are inter winding capacitance in the transformer. You can see the wave form on the MOSFET couples directly to "RTN" on the isolated side and the sharp edges on D6 couples back to the input supply. With out C4 there will be a huge amount of common mode noise on the output.
upload_2016-10-26_7-1-37.png
 
I've never seen 2 before, usually just 1 as in post #7, value can vary too, some wall warts will give you a 'tingle' of you touch the o/p.
There are some import sig gens on ebay that have this cap, and for that reason leak 0.5ma of buzz into your project, possibly blowing it up.
 
Deleted
 
Ron, do you still need the two capacitors if the transformer has a screen, which is connected to earth, between the primary and secondary windings? I appreciate that a screen is not always possible when the transformer primary and secondary windings are inter-wound.

spec
 
The screen helps!

The two red capacitors I added in post #10 are part of the transformer and not added to the PCB. The two caps in #5 are on the PCB.

If you have one screen: There is a primary-winding capacitor to screen. Also there is a secondary-winding capacitor to screen. Then where do you connect the screen. Some times there are two screens. One connected to primary ground and one connected to secondary ground. There will be a huge capacitor made inside the transformer (screen1 to screen2).

I have had good luck making a cancellation winding.
 
The screen helps!
The two red capacitors I added in post #10 are part of the transformer and not added to the PCB. The two caps in #5 are on the PCB.
If you have one screen: There is a primary-winding capacitor to screen. Also there is a secondary-winding capacitor to screen. Then where do you connect the screen. Some times there are two screens. One connected to primary ground and one connected to secondary ground. There will be a huge capacitor made inside the transformer (screen1 to screen2).
Thks
I have had good luck making a cancellation winding.
Can you explain this?

spec

PS: Presumably, the added capacitors would need to be class X or Y (can't remember) failure mode types to comply with safety standards in many countries.
 
When you say screen you mean a magnetic screen, one around the whole outer body of the trans, the kind used when theres a large air gap, yes I spose you'd need to connnect it somewhere.
 
When you say screen you mean a magnetic screen
I was talking about a layer of copper between primary and secondary. Usually 1.25 turns. Not connected so current flows. Usually connected to P or S ground. It is a shield against electrostatic noise. Not magnetic noise.
--------------------------------------------------------------------
cancellation winding
What noise:
Fly back switcher, one primary, one MOSFET, One end of primary at supply (350V), Other end at MOSFET.
Place scope close to the primary of the transformer. One end of the primary is quite. The other end is radiating noise. (sharp edges, 600V in 20nS)

Transformer:
The primary and secondary are two pieces of copper spaced very close together. (definition of a capacitor)
The switching noise cap. couples to the secondary.
If the P to S cap. is 10pF and you add a 10nF capacitor on the PCB from Primary to Secondary (grounds) you then just made a 1000:1 divider that reduces the switching noise on the secondary. (more complicated but....)

cancellation winding:
Start at the cold end of the primary and wind turns. This winding is 180 out of phase with the primary. It will make not-noise. LOL
Noise + (180 Noise) = zero.
For one supply I made a small cancellation winding that way only 1/10 of the primary. Then I added a external capacitor over to the secondary. So with the cancellation winding being 1/10 the external cap. was 10x the internal cap. of the transformer.

There are other ways!
Also the noise is coupling to the core of the transformer, so the fix is a little complicated.
More details if you need it.
 
Yes know what you mean, a non shorted turn of copper sheet.
A shorted turn of copper sheet is what I was reffering to, obviously its placed outside the magnetic flow.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top