Flash Type ADC Digital Voltmeter

705

09:07:59

Here is a great digital volt meter (DVM) based on the flash analog to digital converter (ADC).  Its range is 0 to 10V accurate to 1V.  


INSTRUCTIONS:
Adjust the voltage source in the top left, and the result will be instantly shown on the 7-segment displays in the top right.  This example reads 10V.


DVM Series:
Ramp Type https://everycircuit.com/circuit/4592385207304192
Dual Slope Integrating Type https://everycircuit.com/circuit/6675184624599040
Integrator Type (voltage to frequency) https://everycircuit.com/circuit/6172260362354688
Successive Approximation Type (hex) https://everycircuit.com/circuit/6044286526947328
Successive Approximation Type (decimal)
https://everycircuit.com/circuit/5334865438048256
Ramp Type (negative volts) https://everycircuit.com/circuit/5763215168110592
Ramp Type (auto ranging) https://everycircuit.com/circuit/6110164975091712
Counter Type ADC https://everycircuit.com/circuit/4509930926899200
Tracking Type ADC https://everycircuit.com/circuit/6418942605918208
Tracking Type ADC Full Version https://everycircuit.com/circuit/4688147021299712
Pipeline Type ADC https://everycircuit.com/circuit/5739735974412288
Pipeline Type ADC 2 Bit Stages
https://everycircuit.com/circuit/6081127019970560


GENERAL:
Probably the simplest type of voltmeter to understand, the flash ADC uses an array of opamp comparators to detect the unknown voltage level.  Each comparator is set to trigger at sequential voltage levels created by the voltage divider and the reference voltage. As the unknown voltage rises, sequential comparators will saturate.  A priority encoder of logic gates will translate the comparator outputs to a corresponding digital number.


DETAILS:
In this circuit, I have 10 comparators.  Thus, there is a step size of 1/10 the reference voltage. With Vref = 10V, the step resolution is 1V.  This translates nicely to the displays.  I've marked 9/10, 7/10, 5/10, 3/10 of Vref with waveforms blue, green, orange, red.  Note how the output changes when the input voltage passes the thresholds.  

Vref also sets the voltage that will give the highest digital number (in this case, 10V).

A huge plus for this type is the instant response, no sample rate, and no clock.  The downside is that it is not practical for high bit resolution because of all the comparators required. 

published 2 years ago

Issacsutt

2 years ago

Nice! If only we had more space 😅

592azy2circuitdude

2 years ago

Thanks and yeah, more room would be beneficial. Check out the successive approximation DVM (decimal) in my series for a REAL crunch for space! (If you haven't already)

Issacsutt

2 years ago

Man that’s crazy how long it takes just to display the voltage being read, also really like how you not just explain the waveforms scoped in the sim, but also explain other waveforms of interest! Provides Good insight and is a good habit I think

592azy2circuitdude

2 years ago

I'm glad you liked the description and were patient enough to wait all the way until the voltage was displayed! The simulation really bogs down. But I suppose that can't be helped too much, considering all the components.

Issacsutt

2 years ago

Hmm I wonder it would be possible to use some EC ADCs instead to make a DVM that measures in 10mV increments up to 10 or even 100v. It can easily be done with just 1 ADC for a variety of different increments, 100mV,1v,10v, etc… but only for one place value. I think you would need to do some pretty clever arithmetic to be able to reuse each ADC per level of accuracy beyond the next place value. It would have to reset at each of the tens/ones/tenths places in a kind of counter-like function, but analog.

592azy2circuitdude

2 years ago

You're jumping ahead of me 😅. Yes, I was planning to create a voltmeter using EC ADCs and creating a 'super' DVM. However, I was in this series first creating examples of different ADC/DVM architectures without the pre-made component.

592azy2circuitdude

2 years ago

By the way, I did attempt an auto ranging circuit for a DVM. It's in the series. If you'd like to check it out, let me know what you think.

Issacsutt

2 years ago

Cool concept, but just curious why do you need the ramp reference? Figured its easier to ask than try to dive deep into details first thing

592azy2circuitdude

2 years ago

Are you referring to the ramp signal on the right opamp comparator (orange waveform)? I would say it's more of a time delay rather than a reference. The counters increment until the ramp signal surpasses the input voltage signal. The counters then stop on the number corresponding to the unknown voltage. A Ramp Type DVM style works like this. Does this help?

Issacsutt

2 years ago

Ok I think I see the idea behind it now. At first I was confused why you would want to count up to the voltage level rather than just read it… but now I see that it is just a way to cut down on a lot of ADC inputs. Instead of converting each voltage level into a binary value (which consumes a ton of space and components) the counters just count up to the same value until it’s equal. And it can achieve high levels of accuracy with the only limitation being the number of steps in the counter itself. That’s pretty neat, I’ve never considered an approach like that before!

592azy2circuitdude

2 years ago

Yep, you've got it 👍. That's exactly how the Ramp Type DVM/ADC works. It requires less components than a Flash ADC as you noticed, but the tradeoff is it takes longer time to complete the measurement. There are lots of different ways to make ADCs. If you're interested, you might find other circuits in my series interesting, or an internet search of the topic.

add comment in editor