Electronics > Projects, Designs, and Technical Stuff

TPS40211 feedback resistor - can someone tell explain this to me?

(1/2) > >>

shadewind:
I've already asked this in another thread but it was sort of off topic so I'm giving it its own thread.

I want to build a HB LED driver using the Texas Instruments TPS40211: http://focus.ti.com/lit/ds/symlink/tps40211.pdf

If you look at page 34, Figure 36, there is an application circuit for this. R6 is the current sense resistor and C5, C6 and R4 is the compensation network. R23 is for loop injection when fiddling with the optimal compensation. But what is R13? I'm assuming it has something to do with the compensation but how can I determine its value and what exactly is the function of it? I would be very grateful for answers on this one.

Mechatrommer:

--- Quote from: shadewind on July 06, 2011, 10:18:10 am ---but how can I determine its value and what exactly is the function of it?

--- End quote ---
the value is 30.1K Ohm in page 35. i'm "assuming" the function is to separate high voltage/current from feedback/compensation network. and also r6 and r23 (and probably including the r13) will form current/voltage sense/divider, not r6 alone.

shadewind:

--- Quote from: Mechatrommer on July 06, 2011, 11:14:15 am ---
--- Quote from: shadewind on July 06, 2011, 10:18:10 am ---but how can I determine its value and what exactly is the function of it?

--- End quote ---
the value is 30.1K Ohm in page 35. i'm "assuming" the function is to separate high voltage/current from feedback/compensation network. and also r6 and r23 (and probably including the r13) will form current/voltage sense/divider, not r6 alone.

--- End quote ---
Yes, that is the value for this particular design. But the fact that it's 30.1k 1% and not 33k 5% suggests that it's value isn't really chosen to be just "really big". Also, it's not a divider. R6 is a current shunt resistor for sensing the current through the LED string. If we removed the compensation network I don't see any reason why it would be included since the input impedance of the inverting EA input is large that 30.1k is nothing in comparison.

qno:
Looks to me like a current measurement resistor.
10 ohm will give you 10 mV for every mA.

shadewind:
That's not the resistor I mean. I mean the 30.1k, 1% resistor. It doesn't seem to be needed when using a voltage divider for voltage feedback.

Navigation

[0] Message Index

[#] Next page