complete fix constant voltage for the GSM module

[fatemehjml]

I have one GSM module Quectel M95 in my design. I power it up with one 4V switching regulator. The voltage range can be 3.7-4.2V. When I check my designs with Quectel support they ask me to use one Linear regulator after the switching regulator to have a better more constant voltage for M95 module. As you may know, this module is so sensitive in voltage changing and it will be restarted in any changes of the voltage. the device will be used in vehicle with lots of noises.

Now I designed my switching power supply and I don't want to change the values. However if I want to use a LDO after the switching regulator and have a 4V output, I should change the output voltage of switching regulator and increase it, because the minimum voltage difference between input and output voltage of the LDO should be at least 1V.
Does anybody have a better trick to have a better more constant voltage for M95 module except the LDO after the switching regulator? , or anybody know a LDO which is able to have the exact input voltage in the output?

Thanks in advance.

[robertferanec]

In one old project I have seen problems with GSM module. Problem was not enough capacitors on power supply.

On the GSM module, there may be quite big current spikes when modem is sending data and if environment is changing (e.g. in cars when you travel around areas with good and bad signal). Could not be this the same thing?

What I mean, could not be the main issue not the noise on the power rail, but the spikes created by changing currents? I know it may sound the same, but you still maybe could use the switching power supply. Just be sure the power supply is good enough to cover the high current spikes.

It's just an opinion. I could not download the datasheet (the web wants me to register), so I may be wrong

[oscargomezf]

Hi fatemehjml,


A few months ago, I was working on the proposal of a similar project, and I was researching a lot about this GSM modules. The problem is the burst mode, that it has a large current peak. Eventually, the project was cancelled and I can't check everything I had learnt...

You have to take a look to this IC: Step-Up DC-DC Converters with Precise, Adaptive Current Limit for GSM: MAX1687 | MAX1688. It can be very useful for your application: https://www.maximintegrated.com/en/d...ex.mvp/id/1980

I know there is one cheap EVB to check that everything works fine: https://www.maximintegrated.com/en/p...1687EVKIT.html


Best regards.

[mairomaster]

Don't you have any other voltage on you board which you can use to power the LDO? Like 5V for example?

[fatemehjml]

Finally I changed my design and use a 5V switching regulator, and then two separate 4V LDO and 3.3V LDO for GSM module and processor!

[nachodizz990]

Im designing for an automotive 4G SIM7100 and i suggest you to use a good regulator with fast response and follow the required l and c values for 3.3 and adjust vfeedback resistors to output 3.8 volts, AP6550, then multiply the output caps (parallel) to have at least 330uF I place bootstrap caps and compensation caps C0G to not deviate along time

Note that layout of a buck converter dratically impacts its perormance

[nachodizz990]

This are my modyle caps, place testpoints to check that supply dont falls below mÃminum voltage during a transmission burst

[JardCrocker]

Hi...as per my knowledge a constant voltage supply, whether DC or AC, is ideally a source where the output voltage is constant independent of load. This means zero equivalent internal impedance. Never really achieved, but with electronic circuit control can be very good. It may be possible to build a 10 V DC 10 A supply where the voltage changes less than 0.001 V from zero load to 10 A load.

pcb assembly