LCS708 Stability problem
I follow the DER-385 design with the Pixls design tools for a 12V,25A converter and i got good result and stability for only high current values from 1-2Amps the system are in resonnance between 3 and 8 Amp i got modulation and noise after that from 8 to the max 25A the system goes back to stable state.
I double check the transformer construction , value for the control loop , opto CTRL , layout noise issue , Vref are 3.42V , FB pin value (pin6) are 0.72V , shape of the signal without any load are like AN55 debug section.
Test done with a stable 400V DC power supply and and external 12V for the VCCH both source are stable without noise measurement have been done with current probe and diff prob.
Comments
Hi i already try some cap change on the FB pin from 22nf to 47nf that make a little diff but not enough (example 4A are more stable) i will try on the DT do you have the information where you inject the signal to measure the control loop with a small resistor on secondairy high speed line (38.3K) or by injecting the signal on the transistor side of the opto i will try to reproduce the DE385 gain and phase plot . thank's
When i change DT cap from 4.7nf to 10nf no change in the stability at low current
you can add 50ohm in serise with opto diode side. inject noise on this resistor.
After doing the gain and phase measurement my result are not like the DER-385 crossover freq are not in the good range same think for the phase can you give me some clue on how to adjuste the pole and the zero (i mean the effect of the part on DER-385 ) to start my modification.
EM
Majority of your feedback components are similar to DER385. You can try to remove noise filter cap next to 38.3kohm if you install it.
This capacitor was not installed (with or without) taht didn't make any diff , the system are still unstable between 3-7amps
Then I suspect if there is any component on the feedback circuit has bad soldering or broken capacitors. Do you have your body plot measured? how does that look like?
I take a couple of time the measurement and this is the result that doens't fit your test ans i double check the component no crack or bad value.
| Attachment | Size |
|---|---|
| gain and phase a 1A load , R injection 50ohms (183.56 KB) | 183.56 KB |
| gain and phase 10A (190.97 KB) | 190.97 KB |
| gain and phase 20A (175.13 KB) | 175.13 KB |
as a debug process, you can try to short the output inductor and see if the oscillation is changed. If yes, using a low ESR output capacitor might help this oscillation.
In the full voltage range(90V-264V), do PFC 600W output, I should choose which chip,PFS729EG or?TKS!
Can you tel me what's the pole 1,2,3 and zero 1,2 defined into der-385 i will try to re adjust that
thank's
I got new result when i increase the load resistor on the opto transistor from 2.4K to 4.7K no change when i short the output inductor but i still have problem with the gain on the system you can see the curve
| Attachment | Size |
|---|---|
| Gain and phase (116.73 KB) | 116.73 KB |
could you let us know what is the output cap brand, type and part #? We try to see if the ESR rating of the cap is a factor here.
PFS729EG is ok to operation with high line condition, not down to 90Vac. Also we have latest PFS3 family also provide high lien family.
Model like DER-385
Nichicon : RPS1C271MCN1Gs , 14mohms and 1500uf electro , 20mohms at 2.18A
Hello, telcobridges.
I spent a lot of time with my 250V 1A power supply, and at start point a had the same problems- instability at certain conditions. The big capacitance of output filter capacitor was my biggest problem. See my pics. May be it will help.
Also try to remove RC circuit connected parallel to TL431.
| Attachment | Size |
|---|---|
| LCS708_INSTABILITY.png (136.67 KB) | 136.67 KB |
| LCS708_INSTABILITY_1.png (56.52 KB) | 56.52 KB |
Thank's for the information i will try this and i let you know
from what I can read from your schematic, the control circuit components are good designs following our design examples. Our design examples are not having this modulation issue. Although you mentioned you debug the layout noise, for the sake of verification, you can try increased decap values on FB pin and DT pin to see if it helps. for example increase 4.7nF to 10nF or higher as a starting point.