For room temperature 21*C LM35 showing,

  1. 23 *C while powered by CPU USB port.
  2. 18 *C While powered by a 7amph 12V SLA.

My controller consist of Arduino, adruino wifi shild, and ATMEGA32. ATMEGA32 is attached to the 4 redundant lm35. regulated 5V and GND Power rail are interconnected for all controller.

a. The lm35 are designed as unit gain and double gain by a LM324. b. AREF of ATMEGA32 are controlled by a LM4040. enter image description here

enter image description here

I want to share some of my observation:

  1. If I do another Analogue read on ground as the same manner That I do for data, e.g. For this observation I an using gain=3 for both data and ground, and taking the difference, the result become far better.
  2. When the power is only Battery and no connection with CPU/AC Main line, there is no apparent disturbance on the output.
  3. In the given figure the room temperature are noted for around,

    a. 1 Hr for Battery and

    b. 45 min for USB USB power. Data are communicated over USB serial cable.

enter image description here

  • 1
    \$\begingroup\$ Schematic, please. Use the 7th symbol from the left at the top of the answer window, the one with a diode. \$\endgroup\$ – WhatRoughBeast Dec 12 '15 at 15:44
  • \$\begingroup\$ I have issue on uploading image of my schematics, so add as a link. \$\endgroup\$ – Dibyendu Mondal Dec 12 '15 at 16:35
  • 1
    \$\begingroup\$ Sorry, but I'm not going to register. \$\endgroup\$ – WhatRoughBeast Dec 12 '15 at 16:36
  • \$\begingroup\$ I have added relevant image \$\endgroup\$ – Dibyendu Mondal Dec 12 '15 at 17:45
  • \$\begingroup\$ What voltage is your +5V rail when it's on USB power? I bet it's not the same as when it's on battery. \$\endgroup\$ – Daniel Dec 12 '15 at 17:48

It seems as though you're running both the 5V USB and 12V battery through the same linear regulator. Since 5V doesn't have enough excess voltage to regulate to...5V, it is being dropped to 4.5V (per comments in question)

I suspect that you are seeing a change in the ADC reference voltage (AVCC or AREF) that causes the absolute output of the temperature sensor to be interpreted differently.

You should either make better provisions for creating a stable reference voltage, or use the internal 2.56 reference of the ATMega32.

| improve this answer | |
  • \$\begingroup\$ Actually 5V USB came from Arduino, that dropped little bit, still my AREF from LM4040 gives accurate 4.1 volt, and voltage at sensor is 4.5V. In the case of 12Volt supply, voltage dose not dropped, but temperature showing lower (Right now 23*C while room temperature stood at 26.3 *C) then actual. \$\endgroup\$ – Dibyendu Mondal Dec 12 '15 at 18:34
  • \$\begingroup\$ Actually now that I look at it, why do you have three stages of filters after the 7805, and what are the diodes (D2, D3) for? \$\endgroup\$ – Daniel Dec 12 '15 at 18:41
  • \$\begingroup\$ Right, but the voltage at the sensor on battery is exactly 5V right? \$\endgroup\$ – Daniel Dec 12 '15 at 18:43
  • \$\begingroup\$ 4.96V( or 4.97V) at VCC_5 and at the sensor and at AVCC. Actually I have found this "three stages of filters after the 7805" is redundant but gives stable result when I use its power to drive all PWM along with Arduino and Arduino Wifi shield power. \$\endgroup\$ – Dibyendu Mondal Dec 12 '15 at 18:52
  • \$\begingroup\$ Are you certain that you have AREF selected as the ADC reference source? \$\endgroup\$ – Daniel Dec 12 '15 at 20:53

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.