LM431 oscillation

Hello Liang,

Can you please try to reattach the schematic?
If you use the "use rich formatting" and then attach the image using the paperclip, I believe that will work better.
Or could you describe the setup further? what are you trying to acheive?

Best,
Michael

Hi Liang,

What is the value of the supply voltage? What Iq current are you trying to get through the device?

Chris

+VSUPPLY=7V, Iq at Vout is 150mA, Iq at LM431 is about 2mA

Hi Liang,

How long does the system stay up before the resistor burns up?

Do you see 5V at the output as expected?  Have you probed the voltage directly from the LM431 device? Looking at the Darlington NPN you selected, it looks like the base voltage will have to be about 1.25V higher than the emitter voltage, which would make the device output about 6.25V.  What are the base and emitter voltages?

Chris 

Yes, the 5V output was OK.

We never see this problem in house. There were three units return from field with the burn resistor.

Do you want call me instead of time consuming talk like this? My number is 978-268-3292, located in Boston area.

Hi Liang,

Did you have any luck with the load transient, from 150mA to no load, while monitoring the current through R1? I was able to get a simulation together in TI TINA using a similar part, TL431 model, and a discrete darlington built according to a datasheet I found for the MJE802. The model does show increased current through R1 when the load on VOUT is removed, though it was still about 13mA, which is below the 25mA max (1/16W of 100ohm resistor would be 25mA max). Perhaps there is other behavior in the actual circuit in response to the load transient that gets the R1 current up over 25mA? Could this be something you can characterize in the lab and determine if a higher rating on R1 may be required?

Chris

Hi Chris,

Could you show me your simulation results (curves)?  From the result, are you recommending using a higher rating R1?  I am try to set up a test in the lab. No results come out yet.

Thanks

Liang

Liang,

See the images I have embedded and also please see the attached TINA schematic file.  I have done my best to re-create your schematic.  It uses the TL431 instead of LM431 based on model availability, but these parts are similar.   Download TI TINA from this link  

The results are also embedded in the schematic.

This is a plot of several values vs Rload.  You can see as the load decreases the shunt current (blue) increases.  I understand this plot shows it goes to 12mA, but this is something you definitely need to confirm in the lab.  What is the shunt current as the load goes to zero?   If that current gets to 25mA you will have a power dissipation problem in R1.  Something else I noticed was the relationship between shunt current and R2.  You may want to play with that value as well. That figure is inside the schematic.

Chris

LM431.TSC