Hi all,
I'm working with these boards, NRF24L01+. IKn order to stabilize the Vin, I put a electrolityc capacitor (10uF/25V). The reliability of communication is very sensible to Vin variations, and with the capacitor all works very well.
The problem is I tried a little capacitor 3x7mm of 10uF/16V, and I tested it, but the thing dont works. Put back the 10uF/25V and then works again.

The circuit Vcc is 3.3V, so if the capacitor is 25V or 16V dont sholud matter, isn't it?

Maybe all capacitors buyed are defective?? (1.000units bag) I tried 50units and the problem is constant.

This is the board:
https://es.aliexpress.com/store/pro...lgo_pvid=570535cc-66a2-4472-99ff-0486a2b3e76d

 

hi Pepe,
What type of capacitor are the two types you have tried.?
Electrolytic . Tantalum etc ...
E

 

Electrolityc, both.
they are 20% tolerance, measured with polimeter and:
10uf/25V = 7,6uF
10uf/16V = 7,5uF

 

hi,
If you have a 10uF 16V Tantalum capacitor try that type.

If you do not have a Tantalum use a 10uF Electrolytic but add a 10nF or 100nF capacitor in parallel with the 10uF,
E

 

To connect 2 capacitors in parallel is the same that double the total capacity, isn't it?
It will be same that put a "20uF" capacitor?
But the think I dont understand is why the difference in Vmax for the capacitor makes it work or not.

 

To connect 2 capacitors in parallel is the same that double the total capacity, isn't it?

hi,
Only if the two capacitors are the same value, ie: in your example two 10uF in parallel would be 20uF.
But a 10uF and 100nF in parallel would only be 10.1uF.
The 100nF is better suited for decoupling high frequencies than a 10uF electrolytic.

The difference in the Vmax of your two test caps, maybe due to changes in the way the caps are manufactured.?

Does the 10.1uF combination work with the NRF.?
E

 

Let me try.
I'll put the results here. Thank you very much for your help.

 

Please post a picture of each capacitor.

 

For many voltage regulator stablizing applications you must also have a 0.1 microfarad capacitor connected very close to the regulator device. So a bigger capacitor farther away will not stop the high frequency oscillations. The fact is that the internal inductance of the larger capacitor is the problem. So there may be nothing wrong with those capacitors except that they are not the right ones.

 

With all capacitors, "equivalent series resistance" (ESR) can vary quite a lot from one particular design to another. ESR acts like a resistance in series with the capacitor and is often the thing that sets the minimum impedance for the capacitor. In circuits where there are fast-changing currents of quite high amplitude, ESR can actually be more important than the amount of capacitance in determining performance. In both aluminum electrolytic and tantalum electrolytic types, most manufacturers produce varieties with low ESR and others that look similar or identical that have higher ESR. Often a physically larger capacitor of the same type will have lower ESR than a smaller one. Since higher voltage rating goes with larger size, this also means the cap with the higher voltage rating has lower ESR.

In ceramic capacitors, particularly small surface mount types, there can be a large negative voltage coefficient of capacitance. A capacitor rated at 10 µF at 25 V might be 7 or 8 µF when operated at 10 V but down to 3 or 4 µF at 20 V. This again tends to go with voltage rating and size. A big capacitor will likely suffer from this problem less than a smaller one of the same nominal capacitance.

There are many many different types of capacitors made because "real" capacitors are quite different from "ideal" capacitors and different materials and constructions produces capacitors that have specific characteristics that best suit specific types of applications. A capacitor that is good at audio frequency can be very poor at radio frequency (this is why you often see a small ceramic capacitor in parallel with a large electrolytic capacitor - the electrolytic behaves more like an inductor at high frequency while the ceramic still behaves like a capacitor). One that is good as a high-frequency decoupling capacitor for digital logic circuits can be terrible in the signal path of an audio amplifier because its capacitance changes with voltage or it acts like a microphone - or both. One that is quite reliable when used at 600 volts DC might fail and start a fire if used across 240 VAC mains for noise filtering. Some electrolytic capacitors are useless at low temperature, some are OK. The list goes on and on, and it makes choosing the "best" capacitor for any application a bit of a challenge.

 

With all capacitors, "equivalent series resistance" (ESR) can vary quite a lot from one particular design to another. ESR acts like a resistance in series with the capacitor and is often the thing that sets the minimum impedance for the capacitor. In circuits where there are fast-changing currents of quite high amplitude, ESR can actually be more important than the amount of capacitance in determining performance. In both aluminum electrolytic and tantalum electrolytic types, most manufacturers produce varieties with low ESR and others that look similar or identical that have higher ESR. Often a physically larger capacitor of the same type will have lower ESR than a smaller one. Since higher voltage rating goes with larger size, this also means the cap with the higher voltage rating has lower ESR.

In ceramic capacitors, particularly small surface mount types, there can be a large negative voltage coefficient of capacitance. A capacitor rated at 10 µF at 25 V might be 7 or 8 µF when operated at 10 V but down to 3 or 4 µF at 20 V. This again tends to go with voltage rating and size. A big capacitor will likely suffer from this problem less than a smaller one of the same nominal capacitance.

There are many many different types of capacitors made because "real" capacitors are quite different from "ideal" capacitors and different materials and constructions produces capacitors that have specific characteristics that best suit specific types of applications. A capacitor that is good at audio frequency can be very poor at radio frequency (this is why you often see a small ceramic capacitor in parallel with a large electrolytic capacitor - the electrolytic behaves more like an inductor at high frequency while the ceramic still behaves like a capacitor). One that is good as a high-frequency decoupling capacitor for digital logic circuits can be terrible in the signal path of an audio amplifier because its capacitance changes with voltage or it acts like a microphone - or both. One that is quite reliable when used at 600 volts DC might fail and start a fire if used across 240 VAC mains for noise filtering. Some electrolytic capacitors are useless at low temperature, some are OK. The list goes on and on, and it makes choosing the "best" capacitor for any application a bit of a challenge.

Thank you very much!
Really interesting.

 

Here is a picture of both capacitors used.

 

hi,
Did you try my suggestion.?

Does the 10.1uF combination work with the NRF.?

E

 

hi,
Did you try my suggestion.?

Does the 10.1uF combination work with the NRF.?

E

Still not tested. I'll do in a moment.
A think I tested to, is a fresh 2XAA batteries improve the radio stabilisation.
I'm using a HT7733A to boost up 2XAA to 3.3V
A think I wonder is. What is the impact for a great or little choke in the HT7533A circuit.
I read a little choke produce less stabilisation but can supply more mA, and a big choke (more mH) produce a better stabilisation but with less mA. What do you think?

 

hi,
If you are using a Boost SMPS near to the NRF module it could cause 'noise' problems, ensure you have very good decoupling and filtering on the 3.3v
Why don't you use 3*AA batteries and remove the boost module.?
E

 

hi,
If you are using a Boost SMPS near to the NRF module it could cause 'noise' problems, ensure you have very good decoupling and filtering on the 3.3v
Why don't you use 3*AA batteries and remove the boost module.?
E

To use 3XAA could be a solution, but I'm developing a device where the size matter.
Looking a the Volut form the Boost circuit, with an oscilloscope, should be sufficient in order to see "problems"?

 

The "Arduino" type modules that you are using are known to have problems. Try with a ceramic capacitor 16V/10uF if you can or use the "25V/10uF".

Also if this is an "ESP8266" or other similar type module, you need to connect all power supplies to filter capacitors (4 in parallel of 10uF or 100uF per power pin, depending on the application). Plus use the "Vin" pin only for powering the module.