Voltage Reference LT1236 Output

Go To Last Post
23 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi,

 

I am using an Analog Devices LT1236 Precision Reference (5V) in my AVR project (datasheet). I am supplying 7.5 volts on Vin and receiving 4.71 on Vout. This is my first time using this chip and I have read the datasheet but surely I have not comprehended it all. Since 4.71 volts is well below the margin of error given (0.05%), what could cause this result? I am leaning toward needing to trim the output voltage as stated on page 7 of the datasheet, but I don't fully grasp how to do that. Is that the correct direction to head?

 

Any help or pointers would be appreciated,

 

Kevin

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

This is not General Electronics - this is a very specific question about a particular part - and a competitor's part at that.

 

The Product Page on the Manufacturer's site has links to their support resources, forums, etc:

 

https://www.analog.com/en/products/lt1236.html#product-discussions

 

 

 

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Something does not seem quite right. How are you determining the output of 4.71V? What is your circuit?

 

You're using a precision reference on an average ADC - maybe you should consider using a better ADC if you're wanting better performance?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

I didn't realize this was a forum for Microchip devices only. Sorry for the bother and thanks for the link.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi Kartman,

 

Thanks for the reply, I'm only running through examples to gain a better understanding, the "project" I speak of is nothing grand. I am just trying to learn how to utilize the ADC in general, configuring ADMUX, etc. After connecting this particular chip to the Atmega328P AREF, I was interested to understand how the ADC calculation was functioning, so I did the math and the reference voltage appeared wrong. I then connected it to my scope and my math checked-out. So, even "out of circuit" connected directly to a bench supply, the output was 4.71 volts.

 

No worries, I will investigate over on Analog's site and track it down. I assumed "general" meant "anything to do with electronics within your AVR project". I was wrong :)

 

Thanks for the help,

 

Kevin

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Usually, the causes for a low output voltage are:

 

(1) Input voltage too low. The spec sheet should provide a "dropout voltage". The input should be at least the sum of the output voltage and the dropout voltage. And, preferably significantly higher.

 

(2) Load current too high. That seems unlikely if you are connecting it to the analog reference input of an AVR micro (provided you also have Vcc and AVcc connected). Ahh, you see now, this really does involve AVRs.

 

I have frequently recommended TI switch mode power convert ICs and nobody has complained! And I will continue to do so. Where is the statement disallowing discussion of competitor non-MCU products?

 

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

Last Edited: Thu. May 28, 2020 - 11:46 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

kcloutier wrote:
"general" meant "anything to do with electronics within your AVR project".

Yes: as the name suggests, it's general electronics as it relates to an AVR project.

 

But your question is really a very specific one about the particular component - the AVR really has no bearing.

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kevin, I don't think your question is unreasonable being posted here. 

If you're relying on the scope for an accurate voltage measurement, then you're using the wrong test instrument. The reference chip most likely requires an external capacitor to be stable. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

From the LT1236 Datasheet for the 5V output version:

 

Increase the supply voltage to 10V and see how it works.

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0


That "10V" is just a default test condition - not a specification.

 

There are some tests at other voltages; eg,

 

 

Interestingly, the datasheet doesn't seem to have any direct specification for the minimum input voltage.

 

But this plot does seem to suggest that it hasn't really "settled down" until 10V:

 

 

but, again, these are specific details of the inner workings of the particular chip - so you really need to be going to the Manufacturer with questions about these ...

 

 

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The "chewing out" by awneil in #2 does seem a bit harsh.

Instead he should have given you a good roasting for not posting your schematic.

 

BTW: That's a disguised roasting from me, how can we possibly diagnose your problem given the limited info in #1.

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Go on - surely the obvious & most sensible place to go for specific questions about a particular part is that part's manufacturer ?

 

The fact that an AVR was even involved at all wasn't mentioned until a couple of posts later.

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kartman wrote:

If you're relying on the scope for an accurate voltage measurement, then you're using the wrong test instrument. 

 

This got me thinking...scope probes. How accurate is a 10:1 scope probe's attenuation. I've had a look around and no datasheets mention it.

#1 Hardware Problem? https://www.avrfreaks.net/forum/...

#2 Hardware Problem? Read AVR042.

#3 All grounds are not created equal

#4 Have you proved your chip is running at xxMHz?

#5 "If you think you need floating point to solve the problem then you don't understand the problem. If you really do need floating point then you have a problem you do not understand."

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Brian Fairchild wrote:
How accurate is a 10:1 scope probe's attenuation.

Good enough for an 8-bit scope and good enough to the the difference between 4.7V and 5.0V.

Most folk; professional & hobbyists; buy a DMM first and perhaps a scope second. I wonder why OP didn't double check with his DMM.

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

N.Winterbottom wrote:

Brian Fairchild wrote:
How accurate is a 10:1 scope probe's attenuation.

Good enough for an 8-bit scope...

 

The OP's question just triggered a thought and, as yet, it seems to be a number that is hidden. The probe itself will be nothing more, at DC, than a 9M resistor acting as a 10:1 divisor in combination with a scopes 1M input resistance. 

#1 Hardware Problem? https://www.avrfreaks.net/forum/...

#2 Hardware Problem? Read AVR042.

#3 All grounds are not created equal

#4 Have you proved your chip is running at xxMHz?

#5 "If you think you need floating point to solve the problem then you don't understand the problem. If you really do need floating point then you have a problem you do not understand."

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

OK I had to go and look this up:

Tektronix TPP0101 Oscilloscope Probe | Datasheet

 

Input resistance 10 MΩ ±1.5%

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

N.Winterbottom wrote:

OK I had to go and look this up:

Tektronix TPP0101 Oscilloscope Probe | Datasheet

 

Input resistance 10 MΩ ±1.5%

 

 

Thanks.

#1 Hardware Problem? https://www.avrfreaks.net/forum/...

#2 Hardware Problem? Read AVR042.

#3 All grounds are not created equal

#4 Have you proved your chip is running at xxMHz?

#5 "If you think you need floating point to solve the problem then you don't understand the problem. If you really do need floating point then you have a problem you do not understand."

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

and receiving 4.71 on Vout. ....

Since 4.71 volts is well below the margin of error given (0.05%)

How good is your meter? what brand & model? A cheapo meter can easily be off a few tenths of a volt.  If you want to verify precision amounts you need precise & calibrated equipment!

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

Last Edited: Fri. May 29, 2020 - 06:55 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Input resistance tolerance is not the same as probe factor (AKA attenuation ratio) tolerance. Input resistance is an issue only with quite high source resistances/impedances (typically higher than 1Meg). Probe factor is an issue in all circuits.

 

Tek scopes used to have a calibrator connection (which was sometimes ALSO a calibrator loop for current probes). These were typically 1% in the era when 5% was "pretty good". 

 

Correct, an oscilloscope is generally a poor choice for measuring DC voltages. You can usually do somewhat better using comparison - look at what the scope shows at some known voltage, then look at it at the unknown voltage. Digital scopes usually are able to offset the trace baseline by some known voltage which is then shown somewhere on the display screen. You really DO need to check where that trace sits with zero offset, then apply that same value when you measure a non-zero voltage. All that said, it is also important to do all that at the SAME vertical sensitivity setting so that you do not have screen scale factors also becoming an issue.

 

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Apropos broadcast engineering: because of the critical heights of the various bits of the PAL/NTSC signal (sync tip to black level = 0.300v, black level to peak white = 0.700v etc.) and the general inaccuracy of scopes, the fact that the appropriate signal levels were marked on the graticule wasn't all that helpful without a - tada! - 'god box'.

 

Signal goes into god box; god box adds a calibrated square wave of suitable height to output, scope shows two signals sitting on top of each other. If the appropriate bits aren't aligned, the input signal is wrong (this was from the days when the BBC cared about such technicalities, so unlikely, but that's why we checked) - and if they were, the gain of the scope could be adjusted until the range on the scope was also right.

 

Neil

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Not sure why someone would pretend to use a scope for precision voltage measurements...way too many variables...to top it off, if the wave fills 5 divisions, that's, only 25 ticks to count, each tick is 4% by itself, neglecting all the other errors.  At least if using a digital scope, the visual uncertainty is removed & you can get an 8 (or maybe 10) bit reading.  The specs are slowly improving, however.

These were typically 1% in the era when 5% was "pretty good". 

Remember when 10% resistors were common?  I cringe a little when someone says they are gonna throw some opamps and resistors together, and they "only" need 16 bits of accuracy "for now". 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

barnacle wrote:

...without a - tada! - 'god box'...Signal goes into god box; god box adds a calibrated square wave of suitable height to output...

 

Something like a Cox Video Calibrator?

#1 Hardware Problem? https://www.avrfreaks.net/forum/...

#2 Hardware Problem? Read AVR042.

#3 All grounds are not created equal

#4 Have you proved your chip is running at xxMHz?

#5 "If you think you need floating point to solve the problem then you don't understand the problem. If you really do need floating point then you have a problem you do not understand."

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

That's the kind of animal; most of the ones I saw (pre digital days, remember) came from the designs department at Kingswood Warren, via Calibrations.

 

Neil