Bias circuit values impact on Linkswitch PH constant-current performance?

Posted by: dave-ml on 11/29/2011

Hi,
I have been working on a 230VAC input, 25-45V 700mA output design making use of an LNK419 (no dimming, R pin resistor = 24.9K) and based very closely upon the design given in RDR195. I have found however that my output current is about 70mA higher than expected, but more significantly also varies about another +/-50mA over the output voltage range.

I have also observed that by changing the bypass-pin series resistor in my prototype, the output current and regulation also changes. Generally, increasing the value tends to further increase the output current and further widen the current-range over a the output voltage range! In all cases, I've ensured the BP is being supplied at least 2.5mA to allow for external bias powering.

I have used PI designer and followed its calculations for transformer and component specs
the LNK419 is being used in reduced-power mode, with 10uF on the BP pin
the voltage on the high side of the upper V pin resistors is stable
the current into the feedback pin is well within the CC range of 85uA to 220uA, across the whole output voltage range, and centred on approx 140uA for 35V output
over-voltage protection on the bias-side is not being triggered

Is there something I am doing incorrectly here? Any comments or suggestions for testing/improving the performance of the prototype - especially to 'tighten' the CC range - would be much appreciated!

Kind Regards,
dave-ml

Comments

Hi,

Could you please provide us a copy of the PI Designer files and a copy of your schematic, board layout and transformer construction drawing/details?

This should help us to understand the problem

Regards

PI-Sarek

Hi,
Thank you for your prompt reply! Unfortunately, due to the commercially sensitive nature of the project I am working on, I am unable to provide these files publicly. Would it be possible to send these files directly to PI?

Kind Regards,
dave-ml

Hi,

Please send the files to the following address:

Application.Engineer12@powerint.com

Regards

PI-Sarek

Dear Sarek,

Could you please share the finding, as I face a similar issue.

Thanks!

PI-Troi

Hi there,

The following items can improve regulation;

1- Good coupling between bias winding and secondary output winding. - LNK-PH uses primary side regulation having a tight coupling between bias and output windings is essential. Ideally, bias winding is placed on the outermost layer next to secondary winding and away from the primary winding.

2- Looking at RD195 schematic as reference, place resistor R3 and R6 very close to V-pin and R-pin respectively. Since these pins are signal pins keep the traces and components attached to them away from any noisy nodes with high di/dt or dv/dt such as the drain

3- Resistor R8 in series with the bias diode rectifier D5, is also critical it serves as a filter to any leakage energy coupled to bias winding from the primary winding. Try increasing its value value, but you may need to recenter the set-point of the output by adjusting R6 resistor

4- Lastly, you can use PIXL Design spreadhsheet (latest version can be down loaded from http://www.powerint.com/en/design-support/pi-expert-design-software) optimizing feature for regulation, this section can be found at the bottom part labeled as "FINE TUNING". Also, the PIXL provides two values of resistor called RV2 and RFB2, which you can employ to improve regulation. RV2 and RFB2 is connected across V pin to S pin and FB pin to S pin respectively.

Your comment about ouput current goes up when you increased the value of BP series resistor is an expected behavior. Since the feedback is reying on the voltage from the bias supply, then removing the resistor or increasing its value will draw less power from the bias supply,which will make the voltage goes up and push more current into the FB pin, then the output current will go up.

I hope this helps.