Shop OBEX P1 Docs P2 Docs Learn Events
DeltaSigma ADC in unregulated circuit — Parallax Forums

DeltaSigma ADC in unregulated circuit

bambinobambino Posts: 789
edited 2008-02-10 18:05 in Propeller 1
If I use·an ADC circuit like the one found on the demoboard micrphone, with a propeller powered by an unregulated 3.6V battery(Possibly with diode protect on positive) are the readings I would get going to be valid over the range of battery voltages I'm likely to see.
My guess is it would not work, but thought I would ask!

Comments

  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2008-02-09 20:30
    To the extent that Vdd is within the Prop's operating range, the ADC readings should be ratiometric to the supply voltage. They will not be not absolute. In other words, a reading of 1/2 fullscale will represent Vdd/2, approximately.

    -Phil
  • deSilvadeSilva Posts: 2,967
    edited 2008-02-09 20:43
    Clever of you to make a new thread:

    ANY ADC needs a reference to what "full count" will mean. In fact ADC of whatever kind give you just the feedback that your analogue input is just near the OUTPUT you have generated by your own DAC.

    In other words: Any ADC ist nothing but DAC + Comparator.

    As the voltage generated by linear regulators is +/-2% this has always been the highest accuracy of an uncalibrated Delta-Sigma ADC.

    The situation in the context of a battery discharge curve is worse - you will discharge a Lithium battery from 4.1 Volts downto 3.1 V (= +/- 15%)

    However there are precision reference diodes (LM385 are very popular) you can calibrate with. You can connect them to two "reference delta-sigma pins", or use a - say 25 cents 4051 - analogue multiplexer. The latter will give you an 8 channel ADC (minus 1 for the reference voltage) with 2 + 3 pins only at the Propeller
  • bambinobambino Posts: 789
    edited 2008-02-09 22:07
    @Phil, Thats about what I was thinking. So basically a message recorded at 2.9 volts and output through the speakers on a fully charged battery 3.9 - .3 for diode = singing Chipmunks!

    @deSilva Have you a datasheet on the 4051, Googled, but got a list of ink cartridges and CD players!
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2008-02-09 22:22
    No change in pitch: just louder and, perhaps, distorted if the low record voltage led to clipping due to not enough headroom in the preamp circuit.

    -Phil
  • bambinobambino Posts: 789
    edited 2008-02-09 22:38
    That might be educational to see anyway. I don't have any particuliar use for the micrphone, just some extra pins. I can't stand extra pins!
  • deSilvadeSilva Posts: 2,967
    edited 2008-02-09 23:14
    bambino said...
    @deSilva Have you a datasheet on the 4051, Googled, but got a list of ink cartridges and CD players!
    Funny, they belong to the most popular chips..I am sure you know the even more popular 4066 ?
    www.nxp.com/acrobat_download/datasheets/HEF4051B_4.pdf
    There is also a "stereo" version 2x 1:4 mux = 4052

    However you should use the 74HC... makes of them
  • bambinobambino Posts: 789
    edited 2008-02-10 01:22
    Oh, I thought for a minute you meant either/or a diode and multiplexer.
    I don't get around to using logic chips as much as I like, however I due have the 138A HC series attached. If it would suffice all I would need is the diode of which you spoke right?
    pdf
    pdf
    MC74HCT138A-D[1].pdf
    103K
  • deSilvadeSilva Posts: 2,967
    edited 2008-02-10 08:56
    No,no,no,no! It's an ANALOGUE multiplexer you need smile.gif To my knowledge there is only the (74HC)4051 ff.

    Reference diodes. This is tricky field. You can use Zener diodes of course, but they have bad temperature behaviour and cannot be calibrated. I enclose a link mostly useless to you, as it is a German one, but with lots of interesting details smile.gif
    www.sprut.de/electronic/referenz/index.htm (This site of SPRUT contains much more very knowledgeable things...)

    The author recommends the LM336-2.5
  • bambinobambino Posts: 789
    edited 2008-02-10 15:28
    @deSilva
    I think I will most likely just use a normal microphone set-up(very limited real estate on the pcb), but I must say your idea for an 8 channel adc is facinateing and the referance diode adds a nice touch found in expensive ADCs.
    I can imagine a circuit that would allow for a 4 channel - mius one for referance, but I don't see you theory of 8.
    Even though I will not implement it in this app, could I trouble you to conjure up a schematic for the circuit you purpose?
  • deSilvadeSilva Posts: 2,967
    edited 2008-02-10 17:51 
    
           |
          .-.
          | |
          | |
          '-'
           |         1  2  3  4  5  6  7
           +------+  |  |  |  |  |  |  |
           |      |  |  |  |  |  |  |  |
           z      |  |  |  |  |  |  |  |
           A      |  |  |  |  |  |  |  |
           |   +--|--|--|--|--|--|--|--|--+
           |   |    4051                  |
           |   |               select 0..7|
           |   +--------------------|||---+
                           |        |||
                          .-.       |||
                          | |       |||
                          | |       |||
                          '-'       |||
                        || |        |||
                   -----||-+---+    |||
                        || |  .-.   |||
                           |  | |   |||
                           |  | |   |||
                           |  '-'   |||
               ------------+---+----+++
                        Propeller
(created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)
  • bambinobambino Posts: 789
    edited 2008-02-10 18:05
    So where connecting z to the sigma delta on the propeller and comparing 7 of the signals to the 8th channel connected to the referrence diode. Sweet.
    I do have an app in mind that would need 4 ADC channels, this would definnatly be a lower part count than 4 sigma Delta circuits.
    Less pins too!
    Thank you!
Sign In or Register to comment.