60v power regulator for MCU required

[johnny78]

hi Guys
i need to power an MCU on 60v supply
unfortunately the HV version on lm317 ot 2596 is not available
a friend suggested to connect 2 lm2596 on series for 60v input & use one output
But it didnt work
as a voltage devider to read the 60v dc on analog pin i have used 100k as R1 & 9,09k as R2
using voltage devider calculator & it works But do you have any ideas if i need to use another values?

any suggestion for 60v regulator?

thanks in advance
Johnny

 

[KlausST]

Hi,

i've mentioned before

This may be a disability or laziness. (not meant offending)
In both cases - when you want to continue with electronics design - I recommend to work on your skills and use available aids. This can be:
* handheld calculators
* excel or similar
* simulation tools
* and for sure reading and learning

....even if im an expert I don't need to recalculate every step i work on

No, surely you don´t need to calculate _every_ step but you should "be able" to do so.

or someone else

That´s the point. Someone needs to do this. But in my opinion this is the job of the designer.
Don´t get me wrong. The forum is here to _help_ you to do the basic stuff in future on your own.
If you continously (after about 6 years) still need people to do the same stuff (+, -, *, /) for you then you did not learn anything.
In this case you should consider to pay people for doing your job.

I like to help people, this means that I like the people to learn and understand things, so that they can do it on their own.

Again, it´s just my opinion - and other opinions are allowed:
But Ohm´s law, series connections, parallel connections, voltage dividers, first order RC filters, V, I, R, P should be done by the designer. Even using the formulae given in the datasheets. And Excel could be a big help in this.

Klaus

 

[danadakk]

We want the regulator to stay in its linear region, eg. min Vin - Vout it has
to have, over the range of input V :

From graph, say 25C, we see thats ~ 1.6 to 1.7V

But we also want this to occur over the range on Vin of 40 - 60 V. So
sim shows (solid lines 60 Vin, dashed 40 Vin) :

Sim looks OK. You have to confirm what LM2596 module input C value is.

Note no input R is used. Reason, we want at 40 Vin, 40 Vout. So
design wise we need to account for the needed min Vin-Vout of the LM317
to have it regulate. So we compromise and say allowed input range is
40 + dropout = 40 + 1.7 = 41.7 to 60 Vin is design range. A R aggravates this
input range problem, so leave it out.

All this depends if LM317 is predictable when not in regulation, which
initially takes a small delay (a few uS) to get into regulation, and Vin-Vout exceeds
its dropout V. So in real world, not sim, is there at any time Vin - Vout > 40V
of the LM317 ? We simply can only verify this on bench, for a specific prototype.
SIMs are not 100% accurate due to spice models do not represent the total
behavior of the device. If you plan on manufacturing this, either you examine
a number of LM317s various date codes or you abandon this as a solution.

I still think a simple 20V zener 5% tolerance on input is the right solution here.
Its power level needed = 20V x 1.05 (tolerance) x .075 A = ~1.6W, so a 2W zener
would suffice.



Regards, Dana.

 

[danadakk]

What you can try with the LM317 approach is wire it with an R, set it to 100 ohms.
That will help to insure no transient burns out your test bed/prototype. Look at
LM317 input with a DSO, set trigger level to 40V, single shot, and cycle power to see if
you get any transients at its input. Then reduce the 100 ohms by some amount, say
20 - 30 ohms, and try again. If you don't get any triggers and get to 0 ohms you are
good to go, leaving it out.

There is a 3'rd alternative to protect the LM317, strap a 40V Zener across it,
input to output. That way if any condition attempts to over V it the Zener conducts and
charge is diverted to LM2956 input. I would sim this first just to explore possibility.

Just a thought....


Regards, Dana.

 

[johnny78]

Hi

im done with the regulator section of my project & it works great till now
i want to know how to calculate the voltage drop of a resistor or how to calculate what resistor i have to use before the regulator or zener diode?

now lets go back to the ADC reading
i was reading about the external reference voltage & i have some questions i need your help with

if i have stable 4.9v for my circuit & i have 4.7 zener diode i want to use it as a reference how to decide what resistor i have to use before the diode ?
if i use 4.7v ass reference does it make a problem if my voltage on the ADC pin goes high to 5v?
the atmega328 have 6 ADC inputs i use 2 as SCL & SDA for the I2C devices 1 for measuring voltage & 1 for measuring LDR
so 2 free pins left & i have connected a 10k pull down resistors but later i read there is internal 32k on the adc pins
so what is the best to do ?

any help will be appreciated

 

[KlausST]

Hi,

Resitor for zener:

It´s the most basic formula: (what we call) Ohm´s law.
--> R = V / I
And as always: V is the voltage across the R and I is the current through the R.
So if you have a 4.9V supply and a 4.7V zener .. it should be easy for you to find out the voltage across the R.

Now read the zener datasheet for the zener current.
To this value you have to add the current drawn from this "reference" (I gess it´s neglegible)

And do the calculation for R on your own.

If you still unsure, show us your math.

****
I don´t recommend to use this setup, because:
* I don´t think that the 4.9V is really stable (over time, temperature and load current) ... referenced to the small voltage across the R.
.. and thus the zener current is not stable ... and thus the zener voltage will fluctuate
(I know that the 5V zener has low tempco)
* the initial zener voltage is not very accurate (read datasheet)
* the zener voltage will drift with time

Instead I recommend to use a true "voltage reference" with a circuit according it´s datasheet. In detail it depends on your accuracy and precision requirements.

Klaus

 

[danadakk]

Hi

im done with the regulator section of my project & it works great till now
i want to know how to calculate the voltage drop of a resistor or how to calculate what resistor i have to use before the regulator or zener diode?

now lets go back to the ADC reading
i was reading about the external reference voltage & i have some questions i need your help with

if i have stable 4.9v for my circuit & i have 4.7 zener diode i want to use it as a reference how to decide what resistor i have to use before the diode ?
if i use 4.7v ass reference does it make a problem if my voltage on the ADC pin goes high to 5v?
the atmega328 have 6 ADC inputs i use 2 as SCL & SDA for the I2C devices 1 for measuring voltage & 1 for measuring LDR
so 2 free pins left & i have connected a 10k pull down resistors but later i read there is internal 32k on the adc pins
so what is the best to do ?

any help will be appreciated

What is the power source for the zener ? What V ?

This shows you how to do the calculation https://www.electronics-tutorials.ws/diode/diode_7.html

Note you have to take into load current consideration, as well as Zener Test current spec.
Thats the datasheet current used to spec the Vz.


Regards, Dana.

 

[KlausST]

if i use 4.7v ass reference does it make a problem if my voltage on the ADC pin goes high to 5v?

I don´t know. To answer it I had to read the datasheet. Better you read it.

so 2 free pins left & i have connected a 10k pull down resistors but later i read there is internal 32k on the adc pins
so what is the best to do ?

I don´t know what you want to do / what you are asking for.
Which chapter of the datasheet is the info with the 32k Ohms?

Klaus

 

[danadakk]

if i use 4.7v ass reference does it make a problem if my voltage on the ADC pin goes high to 5v?

You should be fine, but do a test case to see what the ADC reads under that situation.
If its done right it should have saturation arithmetic coming out of the API that is used
to manage and read the ADC. In other words once Vin hits ADC full scale any further V
increase should keep the ADC reading constant at its full scale value. To spec limits of
pin input allowed V range.

Note as I said you should be OK, but compute the worst case delta between Vz and
Vin, over T and V tolerances, if it approaches a Vbe I would be suspicious. For certainty
contact Atmel / Microchip FAE or tech support.

The internal impedance is the input Z of the input signal path. Pulldowns will only affect the
Vsource for the measurement, eg. can it handle the pulldown current and still present its
V to the ADC. Why are you putting pull downs ? Note the input Zin for the analog input
is 100 M typical. Just the ref pin has significant lower Z, see below.

Regards, Dana.

--- Updated May 9, 2023 ---

Dont forget to calc worst case Pdiss of Zener, to make sure you
get the right sized R power rating. Use some margin as well,
eg. if you calc 1W use a 2W R. Again worst case values T and V.


Regards, Dana.

 

[danadakk]

I Think this is a better design procedure to do the worst case design for the
Zener regulator.

Solved Problems On Zener Diode - Electronics Post

Solved Problems On Zener Diode

electronicspost.com

Regards, Dana.

 

[johnny78]

I Think this is a better design procedure to do the worst case design for the
Zener regulator.

Solved Problems On Zener Diode - Electronics Post

Solved Problems On Zener Diode

electronicspost.com

Regards, Dana.

thanks Dana

i need to understand carefully the tutorial you sent

i want to use a shunt regulator as a reference for the ADC so my question is how to calculate the resistor value of RS if i have 4.9v Source & 4.7v zener diode ?

which formula ?
what if i use another lower value is it better as a reference ? i read the mega2560 include about 2.6v internal reference why they use this value?

thanks

https://www.electronics-tutorials.ws/diode/diode_7.html

i think ihave found my answer here
im reading it

 

[danadakk]

Thats too low a input value to make a workable design, eg. the 4.9V.
Because of tolerances, etc.. A 5% zener would worst could be either
at 4.7 +/- 5% = 4.94, which means it would not be in regulation,

You have the V into regulator, use that as source.

This will help. https://www.redcrab-software.com/en/Calculator/Electronics/Zenerdiode-Resistor-Fix


Regards, Dana.

 

[johnny78]

Now read the zener datasheet for the zener current.
To this value you have to add the current drawn from this "reference" (I gess it´s neglegible)

Hi guys
i had time yesterday to read about this article & to be honest its my first time read about zener diodes & when using it this way it draws more current
i've tried to read the ATMEGA328P datasheet with searching everything about Aref but i couldnt find what current the Aref pin consumption is
Actually its my first time need to take care about the current in my projects but after reading i guess i know the reason of alot of problems i had before

So i've decided to use 4.3v zener diode & used this calculator to help me choose right resistor

Thats too low a input value to make a workable design, eg. the 4.9V.
Because of tolerances, etc.. A 5% zener would worst could be either
at 4.7 +/- 5% = 4.94, which means it would not be in regulation,

You have the V into regulator, use that as source.

This will help. https://www.redcrab-software.com/en/Calculator/Electronics/Zenerdiode-Resistor-Fix

& after 2 days of reading & calculating & trying there is something wrong then i used my magnifier & on the zener c4v7 is written
so all the work & calculations was useless
this is my big problem i face here (no reliable shop in my area) once i've got 3w 30v zener diode but when i tried to read the number to search for a datasheet for it Datasheet the zener body was just black & nothing was written on it & when asking the shop manager about this he said its 30v zener isnt it enough?
even the LDR im reading looks like have some problems when its sunny & hot & there is no datasheet to know the temperatue range for its work
sorry for this long story here.

so for my calculations of the Aref voltage & current do i have to take care of Aref pin consumption ?
as i read in the datasheet there is no problem if i used 4.3v reference & get 5v on the ADC pin
But how to know what current this ADC pin consumes ?

thanks in advance
Johnny

 

[KlausST]

Hi,

i've tried to read the ATMEGA328P datasheet with searching everything about Aref but i couldnt find what current the Aref pin consumption is

This is rather straigt forward .. with one small hurdle to pass.
* Open datasheet
* go to "Electrical specifications" or "electrical characteristics" --> find chapter 28
* use the PDF-search for "ref".
--> find "Bandgap reference voltage": --> skip
--> find "refer to": --> skip
--> find "Reference voltage" in the ADC section: (it describes the recommended voltage range for external reference voltage). Your 4.3V are within the recommended range. Good to know.
--> find "internal reference voltage": skip (since you don´t use the internal reference)
--> find "reference input resistance": here you are! You know the input resistance, you know the voltage you want to apply, you know Ohm´s law: --> then you know the input current. This is the hurdle for you. I call it small, because it´s just to apply the most basic electronics formula, thaugt very early in every (!) electronics class.

Side informatiion: There are inputs..
* with "resistive" character, like here. The higher the input voltage, the higher the current. Thus it´s impossible to specify "the input current" in the datasheet.
* .. which are rather high impedance and just draw some stray current. (depends on production). Then the "input current" may be specified , maybe as +/-10nA
* .. which may draw a constant current (rare..) they will specify the current in the datasheet.
* .. which charge a capacitor (like an unbuffered input of a SAR ADC, see Fig 23-8) .. needs an extra thread to discuss
* .. which may have non linear input current. Then usually a diagram shows it´s behaviour
* .. maybe others

its my first time read about

I´m an electronics designer for decades now. Every single day "it´s my first time" doing things. It´s part of the daily job of an electronics designer. My personal opinion: If it becomes too hard for me to "do things the first time" in my job ... I had to stop calling me a "designer". Mabe even quit my job.

Actually its my first time need to take care about the current in my projects

I can´t believe this.
* Never used a LED on a microcontroller? --> it needs to read the LED datasheet for voltage and current. It needs to calculate the current limiting resistor. It needs to read the microcontroller datasheet for outptu current characteristics (loaded output voltage, maximum current). It needs to calculate the current limiting resistor.
(you never experienced that a red LED needs less voltage than a blue one .. and maybe a blue, white, violet one is not even able to properly be driven from a 3.3V supply)

* How did you plan a supply circuit for a microcontroller or other circuitry? Trial and error?

* Never used a transistor?

i guess i know the reason of alot of problems i had before

I´m not surprised with this.

Klaus

 

[johnny78]

thanks Klaus

* Never used a LED on a microcontroller? --> it needs to read the LED datasheet for voltage and current. It needs to calculate the current limiting resistor. It needs to read the microcontroller datasheet for outptu current characteristics (loaded output voltage, maximum current). It needs to calculate the current limiting resistor.
(you never experienced that a red LED needs less voltage than a blue one .. and maybe a blue, white, violet one is not even able to properly be driven from a 3.3V supply)

* How did you plan a supply circuit for a microcontroller or other circuitry? Trial and error?

* Never used a transistor?

actually im using microcontrollers in everything i work on but when using it for control relays or leds using 5A PSU to drive about 500mA its ok i guess (but sure not reliable all times for measuring asap precisely or workin 24h withous stop)
& i never worked with anything needs to use ADC & its really cool to learn all this details i never thought i would need to learn about it
it was my hobby to play with electronics & its Fun for me
anyway thanks for your help & patience

best regards
Johnny

 

[KlausST]

Hi,

My comment about using a 5A power supply.

Many of my microcontroller circuits draw less than 5mA .. sometimes the go to sleep - with supply current in the low microamperes, then they wake up and draw some 10mA.
Back to the power supplies. (Sadly again I recommend to read the datasheets). The datasheets tell you how the power supplies work. Mayn power supplies are specified with a "minimum current". This may be 1% (or any other value) of the specified current. So for a 5A supply the minimum current may be 50mA. If you draw less than these 50mA then the supply may still work, but not like specified.
A known issue of SMPS is the "burst mode" when loaded with very low current (even for 10s of milliseconds). Burst mode means the SMPS internal controller gets activated (to raise the output voltage) then it gets deactivated until the output voltage drops below a certain level. Discontinous operation. You may recognize a "chirping sound".
This is not considered normal operation.
An example: in burst mode a power supply may operate for 2 ms (with it´s normal switching frequency of 100kHz) then it may completely stop switching for 20ms.
Under normal load conditions the same SMPS will continously switch with 100kHz. No audible noise.

During burst mode the output voltage jumps up and down way more than under continous swithcing conditions:
Example: continous switching voltage ripple 10mV (@100kHz), during burst mode 250mV (@40 Hz).

--> Ripple voltage as well as ripple frequency is worse under low load.Overshoot may happen.

Your application may work, but as soon as you do some measurement applications (containing analog signals) or RF applications it may drop the overall performance.

Klaus

 

[danadakk]

so for my calculations of the Aref voltage & current do i have to take care of Aref pin consumption ?
as i read in the datasheet there is no problem if i used 4.3v reference & get 5v on the ADC pin
But how to know what current this ADC pin consumes ?

@OP

The Atmega Vref pin is a nominal of 32K per datasheet, no min or max. So one has
to use some detective and interpretive work to design for worst case.

We do not know the source of what creates this impedance, but in datasheet its labeled as
a resistance, so if we look at pullup pulldown resistance range we see they vary

Fortunately we have the above data over T and V variation, some datasheets only give this
at nominal T and V, making the task even more difficult. Also be aware some datasheets
can label things as impedance and actually mean resistance. Only way to move thru that
mine field is contact a FAE (Field Application Engineer) at vendor and they (hopefully)
will not do an ass pull and just comment but contact design or production or test
engineering to clarify.

So a "safe" bet would ~ +/- 50%. Translate min R then is 16K, which is max current it will need
coming thru zener R. Rest of current thru zener R is for zener to keep it in regulation.

So take max Vref current (Vzmax / Rrefmin) and add Iz test current from datasheet at
Vzmin and I would double that, eg. operate Zener at 2X needed current just calculated.
Should be safe. This is what I consider the "art" part of design, some common sense
wrapped up in experience with a dab of detective work.

As an aside if we ignored ancillary load current, and operated Vz at knee, load current
could then cause the drop across Rz to pull Vz out of regulation, and then operating
point has V at zener to "walk" upward and to right, no longer enough V to operate
the zener. Another consideration is in some design we have to be cognizant of Pdiss
in zener and the dropping R. Here we are only in mA range, low voltages, so only
concern is saving power. But in higher power designs we may actually need power
Rs and heat sink in zener design. Handbook covers all this.


Regards, Dana.

 

[johnny78]

@OP

The Atmega Vref pin is a nominal of 32K per datasheet, no min or max. So one has
to use some detective and interpretive work to design for worst case.

We do not know the source of what creates this impedance, but in datasheet its labeled as
a resistance, so if we look at pullup pulldown resistance range we see they vary

View attachment 183133

Fortunately we have the above data over T and V variation, some datasheets only give this
at nominal T and V, making the task even more difficult. Also be aware some datasheets
can label things as impedance and actually mean resistance. Only way to move thru that
mine field is contact a FAE (Field Application Engineer) at vendor and they (hopefully)
will not do an ass pull and just comment but contact design or production or test
engineering to clarify.

So a "safe" bet would ~ +/- 50%. Translate min R then is 16K, which is max current it will need
coming thru zener R. Rest of current thru zener R is for zener to keep it in regulation.

So take max Vref current (Vzmax / Rrefmin) and add Iz test current from datasheet at
Vzmin and I would double that, eg. operate Zener at 2X needed current just calculated.
Should be safe. This is what I consider the "art" part of design, some common sense
wrapped up in experience with a dab of detective work.

As an aside if we ignored ancillary load current, and operated Vz at knee, load current
could then cause the drop across Rz to pull Vz out of regulation, and then operating
point has V at zener to "walk" upward and to right, no longer enough V to operate
the zener. Another consideration is in some design we have to be cognizant of Pdiss
in zener and the dropping R. Here we are only in mA range, low voltages, so only
concern is saving power. But in higher power designs we may actually need power
Rs and heat sink in zener design. Handbook covers all this.


Regards, Dana.

hi

finally i got c4v3 zener diode But is it right it consumes 250mA?
& when calculating using redcrab software as mentioned

input voltage =5
output voltage = 4.3
load current = 300mA
diode current = 250mA

the result is
series resistor 1ohm
resistor power 385mW
diode power 1w
total current 550mA

why it needs 1w zener diode ?
& is it possible that regulating with zener diode consumes 250mA?

thanks

 

[danadakk]

Where do you get the "needs 250 mA" figure from ?

The Test current, hence regulator Vz, is only 5 mA. Thats the min current you should
design the R for (and include the max current of load).

So use the calculator provided earlier : https://www.redcrab-software.com/en/Calculator/Electronics/Zenerdiode-Resistor-Fix

You use worst case values for all entries, for example Input V min it will be, Vz its max.......to
make sure the R never drops too much V and takes the Zener out of reg.


Regards, Dana.

 

[johnny78]

Where do you get the "needs 250 mA" figure from ?

The Test current, hence regulator Vz, is only 5 mA. Thats the min current you should
design the R for (and include the max current of load).

So use the calculator provided earlier : https://www.redcrab-software.com/en/Calculator/Electronics/Zenerdiode-Resistor-Fix

View attachment 183174

You use worst case values for all entries, for example Input V min it will be, Vz its max.......to
make sure the R never drops too much V and takes the Zener out of reg.


Regards, Dana.

hi
this is what i got when searched for datasheet

--- Updated Jun 4, 2023 ---

Where do you get the "needs 250 mA" figure from ?

The Test current, hence regulator Vz, is only 5 mA. Thats the min current you should
design the R for (and include the max current of load).

So use the calculator provided earlier : https://www.redcrab-software.com/en/Calculator/Electronics/Zenerdiode-Resistor-Fix

View attachment 183174

You use worst case values for all entries, for example Input V min it will be, Vz its max.......to
make sure the R never drops too much V and takes the Zener out of reg.


Regards, Dana.

hi
this is what i got when searched for datasheet

 

[danadakk]

Thats the max you can put thru the Zener. That would occur (not in your case)
when load got disconnected and Zener had to absorb all the current coming
thru Zener Rv in circuit. Thats a calculation you also do on reg circuit design
to see if you have that problem.

Regards, Dana.