Problem facing with TOP244YN based power supply

If you have 12.4 Volts at the control pin then the controller should not be running.  If you can see the DRAIN switching then you doing something wrong in your measurements. 

Looks like you have AC ripple in your output voltage.

You can connect a DC source at the C9 + . Connect the positive terminal of the souce using a diode to block the DC from the Bias winding. In other words, the cathode of the diode should be connected to C9 + side and the anode to the + VOUT of the DC source. Adjust the DC source to 12 volts. This will give you a solid bias winding that it is independent of the switching conditions of the power supply.

Then power up the power supply and check your ripple again. If the ripple disapears it mean you need more turns in your bias winding or increase the value of C9. 

If you still have the low frequency ripple, you can load the power supply to make sure it is running as expected.  

If the power supply is running, then reduce the value of R8.  This will boost your loop DC gain. Now you should have better regulation and reduction of  your AC ripple at the output.  

Regret the delay in responding. I assume there was some problem in the board in which I was working!. I tested the power supply in new version of the board and the power supply was working satisfactorily.  However many thanks for the support you have extended.

I had two more doubts. Though it is out of topic, hope you can clarify them!

1. Of your recent designs, MOVs are not to be seen in the front end!! Can we design the power supplies without that!!

2. If we need to implement soft start, how to do it without much changes in the design!! Also let me know, can we achieve 5-6 seconds of soft start delay.???

regards,

mahesh 

The MOV at the input of the power supply is used to protect the circuit against high voltages surges.   See below document that explains about surges.

http://www.powerint.com/sites/default/files/product-docs/an20.pdf 

 If you do not need to protect your power supply against high voltage strikes, then you do not need the MOV.

For your soft finished circuit, see figure 2 of below link.  http://www.powerint.com/sites/default/files/PDFFiles/der276.pdf

C12, D5, and R15 form a soft finishing circuit. The larger the value of C12, the longer the soft-finishing time. I recommend you substitute R15 for a diode for your capacitor to be immediately discharged every time the input voltage is turned off and Vout goes to zero. 

Thank you very much for the valuable information.

regards,

mahesh 

Further to the query about this power supply design, we are facing another problem, hope you suggest a solution!. The circuit provided earlier is a part of a relay controller unit. Now with the resistive load connected to the relay contacts, while switching individual relays, the power supply section turns off and on. This happens frequently. I assume the power supply is reacting to the electromagnetic noise. The power supply is tested for its full load condition seperately and during that time there is no problem. Only when I am switching the relays, the power supply turns off and on. I have provided a common mode filter, x1 cap, y1 cap as per the schematic, but I have not used snubbers for the output diodes. Please suggest what could be the problem. I have attached the latest schematic.

Could you please send me an sketch of your system? I want to have a better understanding the way it is operating

 Thank you

Victor

Could you please send me an sketch of your system? I want to have a better understanding the way it is operating

 Thank you

Victor

I have attached the system arrangement for your reference. Hope the reported problem is clear. When I am not connecting the load to the contacts of the relay, we have not observed any problem. But once the load ( bulb load) is connected and switching off and turning on the relay frequently, the power supply restarts. We have tried the Power supply alone for its full load, and happens to work fine. For the time being we have used MOVs across contacts of the relay and the frequency of power supply restart has reduced to a minimum. But this is not the final solution i feel. Is this a routing related issue?? or wrong component selection? We have slightly altered the values of certain components. If there is anything else required, let me know please!

I am sorry for my delay in my reply

What I undertand is that  12 V output is used to turn ON?OFF the relays?  If so,   the load of the 12 V is inductive.  am I right?

Do you know if each relay  has a freewheel diode to protect the power supply when the 12 V output is turned OFF?

Absolutely you are right. We are using this 12V to turn OFF / ON the relays, for the same outputs we have used a post regulator and the output of which is connected to the micro-controller.

1. We have used free wheel diode which is built inside the relay driver chip ( using ULN2003).

2. We have tested the power supply alone for its full load (resistive) for its normal operation.

3. We have checked  the unit with the relay interface but without load to its contacts. This works absolutely fine.

4. The unit restarts once in every 15 times the relays are made on/off, when 60Watts ( 15W x 4) of bulb load is connected to the contacts of the relays.

We would like to know, under what noisy (radiated or conductive) circumstances the power supply can  restart by itself???

If you can provide your direct mail ID or skype ID , we can communicate faster and get our problem solved at the earliest as we have reached critical stage about this project  with our customer. We have produced 100 such units and mostly all in all we have this problem.  I am available on this registered mail ID and my skype id is vmholla.

Hope you have seen the layout which I have sent earlier! Do you suspect any layout issues? Do you need any further details, kindly let me know. 

Just to keep you informed, out of the 6 relays we have used, two sets of 2 relays, meaning we have paralled the coil of 2 relays to achieve 2 change over contacts. 

Please do the needful.

regards,

mahesh 

Another two observations:

1. With the use of compensation capacitor, across cathode and reference of TL431, there is ripple at the output. However if the capacitor is removed there would be no ripple at the output. But we can see high frequency noise on pin 1 of the opto-isolator (without compensation cap)

2. The ripple is maximum at minimum load. However with higher loads ( 75% of the rated load), the ripple is minimum.

Hope these points would lead to solutions to the problem we are facing.

regards,

mahesh