I'm trying to build a step up converter to for from 6.5-8.3V DC all the way up to 48V with a maximum load of 15W.

I've gone with SEPIC configuration of an LT8362. I've gone with SEPIC over boost only to allow the output to be switched off completely.
My schematics attached below.

The problem I have is this. At no load Vout is a nice 48V, however as the load on the output goes above 2W, Vout drops back down to about 20V. Is someone able to give me some tips on what I can do to try an diagnose and resolve this problem?

I did try cutting some tracks and removing C3 to switch it over to a boost converter. I had the same problem but at 2.3W rather than 2W.

I was able to use LTSpice to simulate the circuit but the output was stable.

C3, C11	Ceramic Capactior 1206 1uF 100V X7R	Samsung	CL31B105KCHSNNE
C10	Ceramic Capactior 4.7PF 100V C0G/NPO 0603	YAGEO	CC0603CRNPO0BN4R7
C12	Ceramic Capactior 0603 10nF 16V X7R	YAGEO	CC0603KRX7R7BB103
C14	Ceramic Capactior 0603 1nF 100V X7R	Samsung Electro-Mechanics	CL10B102KC8NNNC
R1, R2	RES 1M OHM 1% 1/10W 0603	Stackpole Electronics Inc	RMCF0603FT1M00
R4-R6, R8	RES 34K OHM 1% 1/10W 0603	Stackpole Electronics Inc	RMCF0603FT34K0
R7, R9	RES 0 OHM 1% 1/10W 0603	YAGEO	AC0603JR-130RL
Q2	0.17A Id, 100V Vds, N-Channel MOSFET, SOT-23		BSS123-7-F
C1, C2, C4-C9	Ceramic Capactior 0603 4.7uF 100V XR7	KYOCERA AVX	12101C475K4T2A
U1	Switching Voltage Regulators Low IQ Boost/SEPIC/ Inverting Converter with 2A, 60V Switch	Analog Devices	LT8362EDD#PBF
L1	INDUCT ARRAY 2 COIL 15UH SMD	Bourns Inc.	SRF1260-150M
D1	Diode Schottky 150 V 1A Surface Mount SOD-123FL	SMC Diode Solutions	DSS115UTR
Q1	MOSFET P-CH 30V 8A 6TSOP	Vishay Siliconix	SI3421DV-T1-GE3
F1	FUSE 0603 6A , 70A In rush current	AEM	F0603HI6000V032TM

 

What happens to the battery voltage? What type and size of battery are you using?

 

What happens to the battery voltage? What type and size of battery are you using?

I'm not using a battery for testing. I've got it hooked up to a 300W bench supply running at 6.5V. I put an oscilloscope probe on pin one of the inductor and it was stable at 6.5V.
I've got the output connected to a dummy load set a to a constant power output.
If I changed the input voltage to 8.3V I was able to increase the output load by about 0.1W before the output voltage dropped off.

 

The LT8362 is limited to 2A. You are asking for 15W/6.5V=2.3A even if it were 100% efficient, and there were no current ripple.

 

Thankyou, That may explain it. I don't know how I missed that. I'll order an LT8364 in and see if that fixes things it's a 4A version of the same part.

 

Make sure that you accurately calculate the peak current.
On an ordinary flyback converter, it can be as much as 6 times the average current. On SEPICs it will be less, provided that the input current remains continuous.

 

Just an update to anyone else who has the same issue. I was able to improve the performance by lowering the switching frequency down to 300kHz. This allowed me to get about 6.5W out of the device before the output voltage dropped off. Not the 15W I needed. (The LT8364 had a very similar performance, although not sure if that had something to do with only one SW pin being connected.) In the end I've decided to spin up another design using a switch mode controller with an external MOSFET in a boost configuration to keep things simpler.

 

In SEPICs the current through the MOSFET is the sum of the input and output currents.