Maker Pro
Maker Pro

How to lower voltage??

new_hope

Oct 22, 2011
18
Joined
Oct 22, 2011
Messages
18
I'm trying to power a 4v gsm module with a 5v power supply and need to lower the voltage so i don't mess anything up. I know the simple equations such as V=R*I ect. I need to lower 5v to around 3v...I'm thinking i would accomplish this with resistors in parallel or series to facilitate a voltage drop (i think that is the right term).

I also know the Vt=Rt*I.
So in a parallel or series or circuit combining the two you would find the Rt (total resistance) by adding all the values together. Then you would plug this value into the equation with the known 5V power supply to give the overall current of the circuit.

From there you can go throughout each resistor and discover the voltage drop by filling in the overall current (discovered from our last formula) and the resistance of that current resistor.

So this tells the amount of voltage drop. In a circuit containing multiple resistors each resistor would have a different voltage drop and the voltage drop over all the resistors would add up to the voltage sourced by the main power supply. But would it be safe to use this voltage drop to be a source to a lower voltage rated device (my gsm module).

i believe the correct term is voltage drop....i may be referring to this wrong. Am i on the right path?
 
Last edited:

Resqueline

Jul 31, 2009
2,848
Joined
Jul 31, 2009
Messages
2,848
It's not safe to use only resistors to drop a power supply voltage to a complex module like that. The current consumption varies "unpredictably". Use an electronic regulator.
 

new_hope

Oct 22, 2011
18
Joined
Oct 22, 2011
Messages
18
Here is the datasheet. It recommends a 4v input to the device. http://www.sparkfun.com/datasheets/Cellular Modules/ADH8066.pdf
the total voltage drop of the device during usage is < then or equal to 400mV according to the below link. So a constant 4v should suffice for all usage. 4.3 would be safer for longer power supply lines.

here is the hardware link which again states that the VBAT to the device should be between 3.5 and 4.5 volts. It is on page 7.
http://www.sparkfun.com/datasheets/Cellular Modules/4-ADH8066-HW-Applicationv1.3.pdf

here is the software development guide which is where i got the below. It seems the two links from the same site where i purchased the module from (sparkfun.com) seem to contradict each other. http://www.sparkfun.com/datasheets/Cellular Modules/7-GSM-GPRS-Software-development-guide.pdf

yes your correct the voltage consumption can vary substantially. The datasheet has a high input max, high input min, low input max, low input min. ect. But as long as i have enough constant voltage to power it during its power drops so it doesn't go below 3.5v then i will be good. The max power drop (during transmission) is less then or equal to 400mV. I purchased it from sparkfun before i read the datasheet and didn't know its power consumption would vary so substantially. Other electronics parts are TTL compatible. I figured i would be able to send it some AT commands and get it working off a 5v power supply.

Got any good articles that show implementations of electronic regulators? I am just going to go with the recommended 4v constant supply that was mentioned in the datasheet i first linked. preferably one cheap....also i am needing to learn how interface them anyways with an MCU to control power to other components (like motors).

As long as i just use a resistor to make a drop voltage around 4v -4.3v then i should be safe...correct?. But i would also need to limit the amperage. And im assuming if i used another resistor how would i know if its changing the amperage or the voltage. outside of voltage drops over resistors if you are adding in resistance to a stable voltage source then the voltage will stay the same and the current will go down.

So once i made my voltage drop i would just add in another resistor to lower the amperage to a safe level. Or would the amperage be universally determined in the circuit based on the total resistance as i stated in my original post. I think it would. Now im contradicting myself. I would like to get this thing at least safely powered before the end of the day. That way i can work on my AT commands and my door sensor and light controls. (I'm trying to have it text me when certain doors or windows are open and allow me to control lights).
 
Last edited:

new_hope

Oct 22, 2011
18
Joined
Oct 22, 2011
Messages
18
http://www.radioshack.com/product/index.jsp?productId=2062601&CAWELAID=107596962
could i use this adjustable voltage regulator and do you have links to a datasheet on how to inteface and adjust how much voltage is sent out. This is rather affordable and i could go to the store and pick it up to avoid shipping costs.

http://www.national.com/ds/LM/LM117.pdf I believe this datasheet contains the adjustable regulator mentioned above. If that is the case then the output voltage is set by to external resistors and it can output 1.5v to 37v at 1.5A. So i would have to bring down the amperage with a resistor before connecting it to the gsm module to avoid frying it.
 
Last edited:

Resqueline

Jul 31, 2009
2,848
Joined
Jul 31, 2009
Messages
2,848
A power requirement of 3.4-4.5V and 2.5-350mA (1A max, 2A peak) calls for more than resistors. The spec' suggests two 330uF low-ESR cap's to cover the peaks.
The module draws only the current it needs. A resistor would make the voltage drop (below 3.4V) and the module would cease to work properly. Forget resistors.
These modules are clearly made to run off a lithium cell. You could use a low-drop adjustable linear regulator running from 5V instead. Use diodes if in a rush.
The LM317 is not a low-dropout regulator, it needs at least a couple of volts "headroom" to work with (so neccessitating a 6-7V input, at least).
 
Last edited:

new_hope

Oct 22, 2011
18
Joined
Oct 22, 2011
Messages
18
A power requirement of 3.4-4.5V and 2.5-350mA (1A max, 2A peak) calls for more than resistors. The spec' suggests two 330uF low-ESR cap's to cover the peaks.
The module draws only the current it needs. A resistor would make the voltage drop (below 3.4V) and the module would cease to work properly. Forget resistors.
These modules are clearly made to run off a lithium cell. You could use a low-drop adjustable linear regulator running from 5V instead. Use diodes if in a rush.
The LM317 is not a low-dropout regulator, it needs at least a couple of volts "headroom" to work with (so neccessitating a 6-7V input, at least).

Aren't the capacitors mentioned put in place to remove interference? If you could link the page number and exact link from where you read that. All together all three pdfs are probably a 100 pages. We might be on different pages. specifically the page where you read about the amperage of the device and the 1A max, 2A peak.

So what you are saying is i could use the above mentioned adjustable regulator even though its not low-dropout but that i would have to use an input voltage higher then my 5V wall supply (6V-7V?). if not what is an adjustable regulator that is low-dropout? I don't necessarily agree when you say the modules are made to run off a lithium cell. There is no way for the module to distinguish between the physical power supply source whether it be a DC wall supply or DC supply from a battery or combination of batteries. For home applications it would not be practical or cost effective to purchase a lithium cell and a lithium cell charger to keep the battery charged if it is never going to move. It would be cheaper to buy a 7$ 9V wall supply then regulate it down. but in remote applications (like a homemade cellphone) or applications that have no plug in connection lithium would be the smartest choice because of its inherent benefits over traditional disposable batteries or rechargeable batteries.

And i understand the module draws only the current it needs but they would not place limits on the max input voltage and amp current if it was not going to damage the module. I'm trying to understand correctly exactly how much voltage and max current i would need to send as input to the module (note: not the regulator). As stated before i could get by with 4V (note: for the voltage...not the Amperage) correct? I would just need to reduce the current going from the regulator from 1.5A down to 300mA so i dont fry the module. Unless you are suggesting the module will not allow excess current to fry it. How would the capacitors limit the amperage while leaving the voltage unaffected at the same time they are allowing the device to draw its 2A peak. Do you have links to any articles explaining how capacitors are used in a manner similar to this?

If that were the case then microncontrollers would not get fried when you put to much current up to them....which is not the case because i fried an MCU in the past. Unless if there is overflow protection on this module that i did not read about?
 
Last edited:

Resqueline

Jul 31, 2009
2,848
Joined
Jul 31, 2009
Messages
2,848
Ok, referring to this document/datasheet.:
Page 7 shows the 2A current pulse that the power supply has to contend with.
Page 9 shows a suitable power supply with the recommended capacitors on the output. These are neccessary reservoars to enable supplying the sharp 2A pulses.
Then referring to this datasheet.:
Page 3 shows some current consumption data, page 5 shows the most extreme consumption that can take place, page 9 shows the typical consumptions.

When I said it was made to run on a lithium cell it meant just that, not that it can only run on a lithium cell. Of course it cannot distinguish between that and a PSU.
The power supply voltage is simply tailored to the lithium voltages. If it had been meant for non-mobile power only it would have been made to run on 3.3V or 5V.

The supply voltage range is a requirement, the currents are only information. Think of the voltage as a highway speed limit. Going above or below will have consequences.
The current consumption can be considered analogous to gas consumption.
Your mileage can vary, but it's your own business as long as you make sure you have enough gas to get to your destination - and don't stall out on the highway.
Also, why doesn't your cell phone charger fry when you plug it into the 15A wall circuit? After all it's specified to draw only 125mA..
And when you turn on the tap you only get the water you want, not the whole dam. Why is that so?
Capacitors can be considered analogous to a water or air pressure tank, storing a supply for short bursts that the pump or compressor would be unable to supply.
 

new_hope

Oct 22, 2011
18
Joined
Oct 22, 2011
Messages
18
lol we were on the exact same page i just didn't even look at that upper graph.

well i went back to my apt and looked at my breakout board (yes i never mentioned it originally..i purchased it seperately) for the gsm module and not only does it have a connection for VBAT that i mentioned above but also a connection specifying 1.8V (since it specifies that im guessing the extra 1.8 volts would not pose a threat to the gsm module) Essentially the breakout board makes it possible to solder the connection on the gsm with 22 awg wire. It just gives holes for all the connections mentioned in the first datasheet i linked.

I'm not hooking nothing up to this gsm module until i know its going to be safe. I fried a 2$ mcu and didn't really care cause it was only 2$. This gsm module is 50$. I would rather be safe then sorry and yes im hard headed.

I'm looking at diode tutorials and i understand how capacitors work but do you happen to have any links to how the capacitors allow the voltage to remain unchanged while allowing the current drawn to be such a broad amount. I'm not asking you to link them because i think you are wrong im asking you to link them because i simply dont understand and i wouldn't know how to accuretly word a search to suffice my appetite. Don't capacitors only discharge once they have become full? so until they became full no current could be drawn by the gsm module meaning it could not draw its peak current or any current for that matter.

I'll try to find that page where you saw it requiring 2 capacitors to allow the peak current to be drawn as the diagram i saw had a schematic of how to arrange the capacitors to remove electronic interference and noise so the device would not pick it up. Im guessing since 2 capacitors are used that in the arrangement they are used it would allow one capacitor to give off the peak voltage while the other is empty or partially filled. Im still not seeing how the discharge of the capacitor could be timed exactly to the peak current draw. Unless if the peak current draw is always the same (the diagram on page 7 only showed 2 square waves so it was largely vague) then you could use the formula to determine how long it would take for the capacitor to discharge then have it match up to the peak current draw of the gsm module.

If
 
Last edited:

new_hope

Oct 22, 2011
18
Joined
Oct 22, 2011
Messages
18
okay i see the circuit you are talking about. It is the manufacturer recommended power supply circuit and has 4 capacitors diodes and some other chip. I guess i gotta order all those parts if i want to get this thing functioning. I have never actually soldered a circuit before. In my mind im thinking i have to layout the components and wires exactly as in the schematic but in reality they can be in any physical arrangement as long as the flow is the same (minus a physical arrangement that breaks the recommendation of not having any signal lines over or under the power supply line). Don't be surprised if i ask some very noob questions when i start wiring it together and soldering it lol.

You know where i can purchase all those? i know sparkfun probably wont have those capacitor values for sell because i have searched the site for capacitors before. Then i gotta order the chip in the circuit ect. I dont think sparkfun sells diodes either. I'll have to check.

The thing that confuses me is they have another circuit right under it that has two connections that both seem to go from and to the gsm module. VD_TOP and the ON_KEY lines in the below schematic. This circuit is also required? and the v3.0 is simply a connection to 3.0V they just switched the orientation of the v? Note these capacitors under the schematic are the ones that remove the EMI interferance. But i think the breakout board already has them on there as the breakout board i purchased is designed around the ground specifications listed on page 7. So i think it has the power on reference circuit components listed on page 10. Here is the link to the breakout board it has all the eagle files. http://www.sparkfun.com/products/10497

Was my assumption correct that the breakout board already implemented the second circuit that i mentioned (page 9) and that is why the breakout board takes in 1.8v at one connection. This is turning out to be slightly more complex then i originally thought. I just wanna power this damn thing already lol.
 
Last edited:

new_hope

Oct 22, 2011
18
Joined
Oct 22, 2011
Messages
18
I need your input resqueline. It seems you are probably one of the most knowledgeable people on these forums. I contacted the tech support also from the site i bought it from as they will offer help also. But it will probably be a while before they respond since its a weekend.

another problem i encountered is on page 18, and 19 it shows the mcu as outputting 4v from GPIO pins....when in reality most mcu are ttl so they output 5v. How would i convert down the mcu's 5v digital logic to 4v digital logic without losing the binary, hex, and ascii values that im sending and receiving between the gsm and the avr mcu (ATmega328 to be specific). I'm thinking it would not be as simple as regulating down the mcu's 5v logic to 4v with an adjustable regulator. Also this would be kind of expensive and bulky assuming i had alot of pins connecting to the gsm.
 

Resqueline

Jul 31, 2009
2,848
Joined
Jul 31, 2009
Messages
2,848
I'm not too keen to do all the design details/work here, suffice to say there are many logic level converter/shifter IC's available that'll do a bunch of lines in one package.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
The easiest fix may be a series resistor and a diode to V+. If you're really feeling brave, omit the diode and rely on the internal protection diodes.

Check to see if the device has 5V tolerant inputs -- it may. If so, you don't need to do anything.

Also check that it will correctly handle the TTL low voltage as a logic low level in all cases. You may need to look at the worst case specifications for both devices.

I am presuming that the uC has a very low input current requirement -- That's probably a safe bet.
 
Top