MIC29312 based linear driver

[jib77]

I deleted it because what I had posted was wrong.

I though V- and GND were still common, and they are not.

 

[Jufran88]

Hmm, So I updated my schematic and here's how it looks:

This is based on the current source design by Micrel.

rhd: Have you tested this configuration yet?

 

[rhd]

Nope, I've been away all weekend.

Have you?

 

[Jufran88]

I will on Thursday after my exam. I just want to make sure this time before I have it sent to the board house. :beer:

 

[foulmist]

I am not sure about the EN connecting it to the battery directly.... I think it might need some serious resistor like megaohms in between because the maximum current to EN pin is 75uA maybe 40-50uA is recommended.... I think you might fry it...

otherwise the new circuit looks just about right for current source for these regulators

 

[HIMNL9]

Sorry, i've not seen this thread before, otherwise i had already advised you ..... ANY non-floating reference ICs cannot work like the 317.

About EN pin, that is not the maximum current you must feed it, it's simply the maximum current that the EN pin can draw from any power source (it's indicated, usually, for designs that need to drive it from logics or MCUs) ..... regardless the resistor you place there (or no resistor at all, tied directly to +V), this means simply that the pin will never draw more than 75uA, so you can freely connect it to Vin.

 

[foulmist]

oh ok .. my mistake

 

[rhd]

ANY non-floating reference ICs cannot work like the 317.

Can you explain this a bit more for my benefit?

 

[Jufran88]

Sadly, we found out the hard way it doesn't work like the LM series ICs, but after reading some app notes there is a way to make it a current source by placing a resistor between ground and the ADJ pin. Just need to test it.

I guess that manufacturers have different processes in creating various linear regulator types.

 

[Jufran88]

So, I finally had time to test the circuit and it works!

Finally, after weeks of trial and error its alive!

I supplied 6V and 8.4V and current was stable at about 57mA. (1.24/22 ohm resistor = 56.3mA) The resistor was 5% tolerance which is probably why its off by 1mA.

This works better than my LMS1587 driver since with an increase in voltage was an increase in current.

:drool:

I didn't use any capacitors in testing this because I didn't have any at the time.

Testing:

All values were calculated with a 22ohm 1/2W 5% resistor

Voltage supplied (V):	Test load (V):	Output (mA):
6V	                        4.2V	                57.5mA
8.4V	                        4.2V	                57.6mA
4.1V	                        4.2V	                15mA
		
4.1V	                        2.8V	                56.5mA
8.4V	                        2.8V	                57.4mA

 

[rhd]

Have you tested with higher currents?

 

[rhd]

Also - stupid question about notation:

1) What is the down-pointing arrow between C2 and the Adj pin?

2) Your EN pin isn't connected to anything. To trigger the driver, do you just tie that pin to V+ ?

 

[bluuray]

that arrow is GND.. i think
helps when routing..

 

[Jufran88]

I haven't tested it with higher currents, but I can test it tonight.


1. Yes, that is the ground pin. I used a differnet symbol for this schematic, but it works just the same.

2. I left it available as an option to use the enable feature. You can just solder it together with the V+ pin.

EDIT:

I finished testing it using 1 ohm 2% 2W resistors:

Voltage supplied (V)	      Test load (V)	           Output (A)
 
@ 1ohm
       8.15V                        2.8V	           1.24A
		
@ (2) 1ohm resistors in parallel = .5ohm		
       8.15V                        2.8V	           1.93A
       8.15V	                    4.2V	           1.22A

The IC barely gets warm running at higher currents. I could hold in in my fingers for the duration I tested it. Interesting since I had not heat sinked it to anything. My batteries felt warmer than the IC running at those currents.

 

[rhd]

This is exciting. I'd be curious to know the results with a 4.1V supply, and a 3.5V test load.

Jufran: based on the old PCB (which I have) can you see any way to make it work by cutting certain traces with a knife, and soldering in jumpers manually?

 

[Jufran88]

**@ 1 ohm**
@4.1V with my test load set for 3.5V its outputting 145-160mA. Weird that it the output keeps rising.

@4.1V with my test load at 4.2V it outputs 37mA stable

@8.08V and test load for 3.5V it outputs 1.23A stable

@6.06V, test load at 4.2V it outputs 390mA stable

@6.06V, test load at 3.5V it outputs 635mA stable

**.5 ohm**
@4.1V, test load at 3.5V it outputs 150-161mA. Again output keeps rising.

@4.1V, test load at 4.2V it outputs 37.5mA stable

@8.08V, test load at 3.5V it outputs 1.45A stable

@6.06V, test load at 4.2V it outputs 399mA stable

@6.06V, test load at 3.5V it outputs 751mA stable

*NOTE: All the values I calculated so far may have some errors in them since there is no input, output cap. I just wired the IC with a resistor between GND and ADJ and tied EN with V+

You could try to do that, but I think it would be difficult because the ground plane is common with the LD- and V-

However, I'm already sending in a new design for fabrication and it should arrive sometime around May 13. I'll mail you a set