Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

0-50V 0-10A Variable SMPS

Status
Not open for further replies.
Sorry but no you cant do like that.
Should be Mains ->PFC -> 2TFC -> ISOLATION -> output voltage

the topswitch should be grounded to the same place as pfc/uc3842
 

Sorry but no you cant do like that.
Should be Mains ->PFC -> 2TFC -> ISOLATION -> output voltage

the topswitch should be grounded to the same place as pfc/uc3842

It didnt make sense to me. Why do i need isolation even if i already isolated them with transformer ? Sorry but i didnt understand that why the pfc and UC3844 must share the same ground.
 

you have a mains (50hz) isolation transformer at the input? your schematic didn't show it
 

you have a mains (50hz) isolation transformer at the input? your schematic didn't show it

No i dont have. Do i really need 50Hz transformer in order to isolate between mains and output? I have never read such thing in textbooks. As i read from textbooks, we dont need a 50Hz transformer for isolation. Any transformer will block the electrical connections between primary and secondary. 2nd option is to use optocoupler. Instead of using optocoupler, i did use isolated supplies.
 

as discussed, you do not have isolation for your secondary of 2TFC. This is very dangerous. How are you regulating the secondary? Are you doing it via a bias winding...I doubt it, not at your power level. Its very dangerous, I would stop if i were you and read up on isolated switch mode power supplies on the internet. Please do not electrocute yourself.
 

as discussed, you do not have isolation for your secondary of 2TFC. This is very dangerous. How are you regulating the secondary? Are you doing it via a bias winding...I doubt it, not at your power level. Its very dangerous, I would stop if i were you and read up on isolated switch mode power supplies on the internet. Please do not electrocute yourself.

Is this because of extra isolation requirements for bias windings ? U say i need a 50Hz transformer or optocoupler right ? Any high frequency transformer so will not work for isolation ? I just didnt understand how the current will flow through my body when i touch positive rail of the smps output ? Current is not able to flow through my body since there is no grounded point at output.
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
how are you regulating the output? is it potential divider, opto coupler, HF transformer, bias winding?
 

how are you regulating the output? is it potential divider, opto coupler, HF transformer, bias winding?

There is a voltage divider for both voltage and current set. Flyback converter's VCC output that supplies energy to PWM, PFC and some opamps, has an optocoupler. If you look carefully, there is no grounded point at UC3844, voltage and current control circuit and output. So i dont need extra isolation. VCC output of Flyback converter is not grounded. Thats why, when i touch the one rail of SMPS output, there is no chance to flow the current through my body to ground.

- - - Updated - - -

Voltage and current set circuit is almost the same as below.

**broken link removed**
 

the uc3844 chip Is (or rather, "should be") grounded to the same place as the pfc......then it's output should be isolated from that.
 

the uc3844 chip Is (or rather, "should be") grounded to the same place as the pfc......then it's output should be isolated from that.

Oh i have just noticed that i had used the same ground with pfc. That 10nF cap is connected to the pfc ground. Yea its wrong but when i correct these kind of ground faults, there will no need for isolation anymore.

I fixed the ground problem around the UC3844. The 2.2nF cap between primary and secondary at flyback converter is for extra EMI reduction. This is called y capacitor i think. Maybe thats why you thought that isolation is needed. I always see this cap on the internet.

- - - Updated - - -

I just didnt understand because you didnt give me a logical answer about why the UC3844 and PFC IC need to have same ground. There are plenty of examples about seperate grounds on the internet i guess. When u give me a logical answer, i will accept your opinion :)
 

Attachments

  • Schematic.jpg
    2.8 MB · Views: 159
Last edited:

I just didnt understand because you didnt give me a logical answer about why the UC3844 and PFC IC need to have same ground
The uc3844 feeds from the pfc. It uses its output caps as its input caps, they need the same ground.
 

The uc3844 feeds from the pfc. It uses its output caps as its input caps, they need the same ground.

No, sorry, you alwas gave me logical answers but this is not. They dont need the same ground because PFC and UC3844 are completely isolated. Flyback converter giving VCC output feeds the UC3844. I have shown you 2 different working schematics. So it works.
 

topswitch might give supply to uc3844, but I am sure you realise that your topswitch cannot supply your 2TFC. your 2tfc draws power from the output of the pfc.
 

topswitch might give supply to uc3844, but I am sure you realise that your topswitch cannot supply your 2TFC. your 2tfc draws power from the output of the pfc.

Topswitch is already not for supplying for 2TFC. If u look carefully, only power stage of 2TFC is connected to PFC. Topswitch is just for supplying voltage to ICs and fan. Fan is connected to the PFC stage. 2TFC is gonna draw the power from PFC stage. You are confused little bit :)
 

I will look again, my screen is only small, and when I view I either see all in one, but cant see any detail, or really close up, but just a few components visible.
I also noticed the FB and COMP of uc3844 are shorted. That doesn't look right.
 

Why dont u download the schematic? You will be able to zoom if u download. FB and COMP pins must be shorted since LM324 external opamp was used. So i shorted pins and the internal error amplifier became voltage follower. This is the common configuration when we will not use the internal error amplifier. Feedback network was put around the LM324 not around the internal error amplifier.
 

yes, but to use the internal error amp as a follower, you need to connect into the non-inv input, and we have no access to the non-inv input of the internal error amp
 

Dunno if this was mentioned before, but you're using the same ground symbol all over the place. On your safety earth, the negative side of the PFC bus, negative output of the bridge rectifier on the IC supply page (which would short out the rectifier), ground of the control circuit.... until you clear that up I can't even comment on the isolation issues. In general the only part of your circuitry which should connect to earth ground is the mains protection stuff, which you've done fine. But nothing else, otherwise you will get severe safety and EMI issues.
 

Yea, i realized that later but it should work anyway. This way must disable that 2.5V reference pin.

I am not using the same ground symbol. One is classic earth symbol the other one is named as GND. There is nothing wrong about ground, earth or whatever i think. I dont know whether you are able to see whole schematic in detail. I used many transformers to isolate the control and power circuitry. Thats why, the control circuitry has GND symbol whereas the power one has classic earth symbol.

Guys i am not manufacturing it. So it doesnt have to be professional one.
 

Yea, i realized that later but it should work anyway. This way must disable that 2.5V reference pin.
It definitely won't work if you actually connect the safety ground. You do know that inside the AC mains, earth and neutral are tied together at some point, right? This means that at low frequency and medium power, neutral and ground are effectively the same potential. So if you earth the output of a bridge rectifier, it will also be tied to neutral (through a lot of wiring), and this will at the very least blow the rectifier or your current sense resistor. Maybe a fuse or breaker first, if you are lucky.

I am not using the same ground symbol. One is classic earth symbol the other one is named as GND. There is nothing wrong about ground, earth or whatever i think. I dont know whether you are able to see whole schematic in detail. I used many transformers to isolate the control and power circuitry. Thats why, the control circuitry has GND symbol whereas the power one has classic earth symbol.
I see the different symbols. The basic solution is to make another ground symbol are replace every earth ground symbol in your schematic (except for the one actually tied to the earth from the AC mains) with that new symbol. Those two grounds will be isolated from each other.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top