how to get a stable 5v reference

[andywatson]

I need to generate a precise 5v on my Arduino Uno board.  I know the onboard regulator isn't very precise, and I've measured too much variation between the 5 different Uno boards I have.  So how should I go about it?  It seems like most precision voltage references need at least 6v so it can regulate it down to 5v.  So do I need to use a DC-DC boost converter to boost the onboard Uno 5v up to something like 9v and then use a reference IC to generate the precise 5v?

[CrossRoads]

Why not just get a switching 5V regulator? Then you won't see any variation.
http://www.dipmicro.com/store/DCA-0510
http://www.dipmicro.com/store/DCA-0520
Or get 6V, 7.5V if  you want to regulate it down yourself. Maybe use an LM317 with a trimming resistor (10 turn potentiometer) to get it right at 5v.

[MarkT]

Standard approach to this is take a sufficiently accurate voltage reference chip (and maybe a precision op-amp to scale/buffer the output).  Won't give a power supply but will be a voltage reference for an ADC for example.

If the application is analog->digital conversion there are many ADCs with built in references.

But without knowing the specification can't really suggest any concrete examples.

[andywatson]

I need to power some pressure sensors that require a stable 5v excitation voltage.  The sensor outputs an analog voltage that's proportional to pressure being exerted on the sensor.  Currently I'm using the the ADC of the ATmega328 to sample the sensor output.  Eventually I'll probably use a separate ADC chip because I want better than 10-bit resolution.  Each sensor draws about 5ma, and I could have up to 5 sensors on one board.  So if I could find an ADC chip that provided a 5v reference capable of supplying enough current and could read a voltage from 0 to 5v, then that would solve my problem.

[MarkT]

Are the sensors not ratiometric?  By this its meant that the ratio of the output voltage to the supply voltage encodes the value, not the absolute output voltage.

[andywatson]

yes, but I'm using a precision 3.3v regulator as the external reference voltage for the Atmega328, so if the 5v excitation voltage varies, the output of the sensor compared to the stable 3.3v varies as well.

[CrossRoads]

So use  the 5V 1A switching regulator, will be a very stable 5V source for your 25mA plus the arduino's 30mA.
Use the analog devices SPI ADC that's been posted previously with 6 sample & hold inputs & get 6 conversions at the same time.

[andywatson]

So use  the 5V 1A switching regulator, will be a very stable 5V source for your 25mA plus the arduino's 30mA.
Use the analog devices SPI ADC that's been posted previously with 6 sample & hold inputs & get 6 conversions at the same time.

Do you happen to know the part number of that ADC?  I've been searching for days and haven't found one with 0-5v input and multiple channels.

[CrossRoads]

Might have been this one
http://www.analog.com/en/analog-to-digital-converters/ad-converters/ad7656/products/product.html
You can search/filter at analog.com to find others also.

[andywatson]

That looks great, thanks!  There are so many ADC chips out there that I've had a hard time trying to figure out which one to use, but that one seems like a good one to start with.

[retrolefty]

I've had my eye on this one for awhile, but holding out so far.

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=390106010150&ssPageName=STRK:MEWAX:IT

And another:

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=230619695873&ssPageName=STRK:MEWAX:IT

[dalpets]

Why not just get a switching 5V regulator? Then you won't see any variation.
http://www.dipmicro.com/store/DCA-0510
http://www.dipmicro.com/store/DCA-0520
Or get 6V, 7.5V if  you want to regulate it down yourself. Maybe use an LM317 with a trimming resistor (10 turn potentiometer) to get it right at 5v.

Hi,

I am having trouble with excessive time spent in the reflow zone of my sketch. Have tried manipulating zone temps & timings without success. I have come to the tentative conclusion that I need to regulate the 5v supply.

I am having trouble  understanding how the power supply you recommend fits into the scheme of things, given that I need the usb power to run the Arduino Uno setup & monitor etc. You may want to view my OP topic: "Arduino code not giving target temperature" (in Project Guidance).

Thanks for any help

[Wawa]

This post?
https://forum.arduino.cc/index.php?topic=394842.0

No time to read through all those posts, so I might give the wrong advice.

I gathered from your first post that you want to measure the temp inside a reflow oven with a themocouple and AD595. And that you have temp stability problems.

You basically can't measure temp or voltage accurately with default Aref (5volt).

But Arduino has a perfect inbuild reference voltage (1.1volt Aref).

All you have to do is call this reference voltage in setup, drop the output voltage of the AD595 with a voltage divider to <1.1volt, and change the temp converter maths line from 5.0 to ~1.1volt.

I might read all those posts later, and correct this one if needed.
Leo..

[Boardburner2]

Summary.

OP is building a reflow oven to an instructable which is no longer available.

He can take pix and use fritzing but i am not sure if he can understand schematics.

He needs a voltage reference and possibly a divider on the input if its not 5V.

Better code as well i think, away from my PC at the moment so am not much help.

[Paul_KD7HB]

yes, but I'm using a precision 3.3v regulator as the external reference voltage for the Atmega328, so if the 5v excitation voltage varies, the output of the sensor compared to the stable 3.3v varies as well.

You have not given the error rate of the pressure sensors. There is no reason to use a voltage source more precise than what you can expect from the sensors.

Paul