Voltage Regulator MP2329 Question about Voltage Range

Thread Starter

trader007

Joined Feb 27, 2010
249
I need to make this work with a VIN range of 6.5v-24v DC.

In the datasheet on the last typical application circuit, (note 10), it that states that if your VIN is lower than 7.5v, the EN resistor needs to be a different value. Does this only apply to the last typical diagram (the one with 1v out)? Or is this true for any output?

https://www.monolithicpower.cn/cn/d...Datasheet/lang/en/sku/MP2329/document_id/3934

My circuit will have a pic chip, so if I have to I can program that to change the EN value depending on voltage detected, but I'd rather not have to do that.
 

Wolframore

Joined Jan 21, 2019
2,195
It sounds like enable has a limit 4.5v and threshold of 1.3v. You might want to use a Zener to limit the max voltage or set up your divider to fall within the range. The divider uses less current.
 

ronsimpson

Joined Oct 7, 2019
1,125
If you do not want to use the functions of EN; then a pull up resistor and a 5.1V Zener.
Many times the resistors are chosen for low voltage shutdown. You can choose the two resistors to shut down below 6V. Then because the 24V will pull too high add the Zener.
 

Thread Starter

trader007

Joined Feb 27, 2010
249
Thanks for the replies.

Of course, after another day of searching mouser, I find a chip that is better suited for my needs. Wide range input without any restrictions that I can see. This chip is pretty amazing http://www.vishay.com/docs/75921/sic437.pdf

It's crazy how much time i can spend on mouser and still find new things. Digikey too, because their search is actually bit better for certain items.

One question I still have though, do you guys think this chip will be OK for driving LED's? I will have a current regulator, but need to keep the voltage within a very tight range (2.7-2.8v). If I go above of that range even just a little, the amp draw skyrockets. I'm just curious to how stable the voltage output is between a low input and high input (say 6v in or 24v in). The temperature charts show it's stable throughout it's operating range so I'm not worried about that. Just don't have a lot of experience with these types of regulators.
 
Last edited:

crutschow

Joined Mar 14, 2008
26,069
do you guys think this chip will be OK for driving LED's? I will have a current regulator, but need to keep the voltage within a very tight range (2.7-2.8v).
If you have a current regulator, the why do you need such a tight voltage value?
You can drive LEDs directly with a constant-current power supply, not a constant-voltage.
 

Thread Starter

trader007

Joined Feb 27, 2010
249
If you have a current regulator, the why do you need such a tight voltage value?
You can drive LEDs directly with a constant-current power supply, not a constant-voltage.
Because I don't know what I am doing :p

I am planning on using these LED drivers. https://www.digikey.com/product-detail/en/diodes-incorporated/PAM2808BLBR/PAM2808BLBRDICT-ND/8545941

They only take 6v input, and since I want 6-24v input, I thought I needed to get the voltage right first. Am I wrong?
 

BobTPH

Joined Jun 5, 2013
2,757
.What are you trying to do? What is this 6 to 24V input, and what are you trying to drive from it. It sounds like you have come up with a baroque solution to what might be a simple problem.

Bob
 

Thread Starter

trader007

Joined Feb 27, 2010
249
These are the LED's im using https://www.mouser.com/ProductDetail/OSRAM-Opto-Semiconductors/GW-CSSRM3PM-N6N8-XX55-1-700-R18?qs=/ha2pyFaduj0ChcJSjXooXgjzoXpFbdy0Nad9QhbCiJUZID/2gNct%2B9sjupTNfnaHzR63o7Fhm4=

They show the voltage tolerance is fairly specific, if you want to get optimal amp draw per lumen. Since this will be a battery powered device, I do have that concern, so I want to keep the voltage between 2.7 and 2.8v per diode. I plan on running two diodes in series.

the 6-24v input will be 6.5-8.4v from the batteries depending on charge, and 12v-24v from various sources, like automotive or dc power supplies. The higher voltage source will be used while the batteries are charging.
 

ronsimpson

Joined Oct 7, 2019
1,125
LEDs are current devices not voltage devices. It is not good to force the voltage to 2.80. It is better to make a regulator that will hold the current at 1.7A and just let the voltage to be some where from 2.70 to 3.20V.
From the data sheet; Vf 2.70 min./2.80 typ./3.20 max. The voltage varies from one batch to another. Changes with temperature.
1587435458554.png
Because you used MPS her is an option. MP24833-A It regulates current.
There are many "buck" LED pwm drivers. TI has some.
----edited----
What you are trying to do is reasonable. Don't let then scare you away. At one time there were flash-light/torch bulbs for 1.5V, 3.0V, 4.5V, 6V …….12V...24V etc. I made one that worked from 3.5 through 30V and it was very popular.
Heat sink the LED if you want to run it at full current.
 
Last edited:

BobTPH

Joined Jun 5, 2013
2,757
There are plenty of inexpensive buck regulators out there that have current limiting and can handle that input range. No need to use two regulators.

To use them, you would set the voltage to something higher than needed and the current limit to the desired value.

Bob
 

ronsimpson

Joined Oct 7, 2019
1,125
LED power supplies: buck, current regulated, wide input voltage. Some good reading.
MP2480, AL8843, ST1CC40,TPS92512
I randomly picked some ICs.
 

Thread Starter

trader007

Joined Feb 27, 2010
249
There are plenty of inexpensive buck regulators out there that have current limiting and can handle that input range. No need to use two regulators.

To use them, you would set the voltage to something higher than needed and the current limit to the desired value.

Bob
Ok thank you. It makes more sense now. I am going to look for a better suited driver and will post back with what I think the calculations should be. I really appreciate the help guys.
 

Thread Starter

trader007

Joined Feb 27, 2010
249
LED power supplies: buck, current regulated, wide input voltage. Some good reading.
MP2480, AL8843, ST1CC40,TPS92512
I randomly picked some ICs.
That AL8843 looks like the best option for my purpose. It's questionable on the low end if i'm trying to get 5.4v Vo with only ~7v input, but that appears to be the case with all of the chips.. I think if I make Rset 0.082 and L=33uH it will probably give me close to 600ma per diode (2 diodes @ 2.7v each in series for 1200ma total).


2020-04-23_07-52-01.png

I was also looking at the PAM2863, but the datasheet is confusing. The application schematic looks flat out wrong,

https://www.diodes.com/assets/Datasheets/PAM2863.pdf

and you can confirm that by looking at it's evaluation board's schematic that looks wired different but correct, but even that shows two diodes in parallel which isn't quite right either.

https://www.mouser.com/pdfDocs/PAM2863EV1UserGuide-R1.pdf

But on that evaluation sheet, it's also a bit hard to tell what the chip will do at 7v input and 5.4Vo. I might get both chips and just see what ones performs better...
 

ronsimpson

Joined Oct 7, 2019
1,125
but even that shows two diodes in parallel which isn't quite right either.
LEDs in parallel is not good. I have done this with three LEDs. We had to get the factory to sort the LEDs by voltage. One week the LEDs would all be 3.0V and the next week they will all be 3.3V.
1587649689490.png
 

Thread Starter

trader007

Joined Feb 27, 2010
249
LEDs in parallel is not good. I have done this with three LEDs. We had to get the factory to sort the LEDs by voltage. One week the LEDs would all be 3.0V and the next week they will all be 3.3V.
View attachment 205172
Yes that is what i have learned, and it makes sense. I did go to school for fundamental electrical theory, so I get the basic physics of all of this. It's just another step to learn what common practice is and what the available chips tend to do.

But that schematic is wrong both ways! the LX pin is Vo and it's connected to the wrong side of the LED diodes... then the diodes are in parallel. It doesnt make sense and it's directly from the datasheet!
 

ronsimpson

Joined Oct 7, 2019
1,125
it's connected to the wrong side of the LED diodes
Inside the part IX is the Drain of a MOSFET sitting on ground. Current flows to ground.
VIN to Isense is the input to a amplifier looking at voltage across the current sense resistor.
How does it work: (it is upside down compared to many buck PWMs. The load (LED) sits on Vin not Ground. The MOSFET sits on ground not on Vin. Vout is Vin-Vleds. Vout is the voltage across the LED measured as if Vin is your "ground".
When the Switch is on: current flows in the solid blue lines. The voltage across L1 is Vin-Vresistor-Vleds.
When the Switch is open: the switch voltage flys up and D1 holds it to Vin. Current flowing in L1 must continue so it flows through D1. Dotted blue line.
Results: (L1 inductance should be large so the current never drops to zero.) Switch=closed, current ramps up. Switch=open, current ramps down. Current continuously circles through the LEDs and L1.
1587663479314.png
The voltage from LED_bottom_end to ground is not regulated. =(Vin-Vled)
 

Thread Starter

trader007

Joined Feb 27, 2010
249
Inside the part IX is the Drain of a MOSFET sitting on ground. Current flows to ground.
VIN to Isense is the input to a amplifier looking at voltage across the current sense resistor.
How does it work: (it is upside down compared to many buck PWMs. The load (LED) sits on Vin not Ground. The MOSFET sits on ground not on Vin. Vout is Vin-Vleds. Vout is the voltage across the LED measured as if Vin is your "ground".
When the Switch is on: current flows in the solid blue lines. The voltage across L1 is Vin-Vresistor-Vleds.
When the Switch is open: the switch voltage flys up and D1 holds it to Vin. Current flowing in L1 must continue so it flows through D1. Dotted blue line.
Results: (L1 inductance should be large so the current never drops to zero.) Switch=closed, current ramps up. Switch=open, current ramps down. Current continuously circles through the LEDs and L1.
View attachment 205188
The voltage from LED_bottom_end to ground is not regulated. =(Vin-Vled)

OK thank you. I was looking at this late and tired. The evaluation board's schematic shows the same thing... but one question.. the 0.05ohm resistor... how does ~3a of current flow through that? it's only a 1/8w resistor, so the current is actually going through the ISense pin?

https://www.mouser.com/pdfDocs/PAM2863EV1UserGuide-R1.pdf
 

Thread Starter

trader007

Joined Feb 27, 2010
249
No current in the ISense pin. (actually 8uA)
From data sheet: 0.1 ohm = 1A (math=0.1V on the pin) 0.05 ohm = 2A (0.1v)
Watts=VxI, 0.1v X 2A=0.2 watts. or I^2 x R (2A squared X 0.05 ohms = 0.2 watts.
I'm still missing something. I understand this works by a balance of rise/fall between the voltage and amps, but how do you get a nominal 2 amps across the LED's when everything is going through a small (though very little resistance) resistor? I need to think about this more, I know I can understand it... It's just not quite clicking in my head. I need to read your explanation in post #17 about a hundred more times ;)

In any case, I decided to go with the AL8843 as it's a newer chip and from what I can tell it will be more efficient at lower Vin.

Another question I have.. the AL8843 is rated for 3 amps. I selected everything to handle that (diode, inductor), but since I plan to use a PIC chip to control the PWM input, do I need to keep the current sense resistor close to what I want for max output (2amps)? In other words, if I use a .033 resistor to make it 3 amp output, can I simply program the PWM signal to not go above what ends up being 2 amps? I am worried if I do this, it might have a negative effect at lower Vin voltages (like when the batteries are running low around 7.5v).

Sorry if my schematic is a little odd on the layout. I plan on making a PCB and I am importing parts from mouser... I don't have all the symbols entered in yet.
 

Attachments

Last edited:
Top