DC to DC power supply overheating... why?

[steveh]

I'm using the following "inexpensive power suppy module" to provide variable DC voltage of approx 1.75-2.25 VDC to a short length of nichrome heating wire with a cold resistance of 4ohms. It draws about 1/2 amp (ohms law), well within the operating range of the module.

I have a 6V 2.5 amp "wall wart" supplying power to the module.

PROBLEM: The 2596s chip in the module gets extremely hot and shuts downor goes unstable. Is this variable DC module suitable for the application, and if not then what is the best way to supply 1.75 - 2.25 VDC to the heating element?? Cost is not really an issue. It has to be powered by an AC-DC wall wart.

Any help or comments greatly appreciated.
Thanks,
Steve
---------------------

POWER SUPPLY MODULE SPECS:

LM2596S DC-DC Step Down Adjustable Power Supply Module
1. Input voltage: 3.2-40V
2. Output Voltage: 1.5-35V (adjustable) ,input must least high 1.5V than output
3. Output current: rated current 2A, maximum 3A (need to add heat sink), if the output power greater than 15W, combined heat sink is proposed.
4. Conversion efficiency: Up to 92% (output voltage higher, the higher the efficiency)
5. Minimum dropdown voltage:1.5V
6.Voltage regulation: ± 0.5%
7. Dynamic response speed: 5% 200uS
8. circuit protection: SS36
9. Operating temperature: Industrial grade (-40 ℃ to +85 ℃) (output power 10W or less)
10. Load regulation: ± 0.5%

 

[davenn]

hi steve

is this a purchased circuit module or is it one you have built up ?
if one you have built up show us pics of the construction sharp and well lit
show us the circuit diag you used

cheers
Dave

 

[steveh]

Hi Dave, it is an inexpensive purchased module like the one shown here:

http://www.aliexpress.com/item/LM25...able-Power-Supply-Module-13093/598782326.html

 

[davenn]

DC voltage of approx 1.75-2.25 VDC to a short length of nichrome heating wire with a cold resistance of 4ohms. It draws about 1/2 amp (ohms law), well within the operating range of the module.

Yup that sounds OK.... but just for kicks ... how about measuring the current in the output.
I wonder if the nichrome wire drops significantly in resistance when it gets hot ?
(been years since I used that stuff ---- probably High School Physics days haha)

that may be giving you much more current drain than expected ??

Dave

 

[duke37]

The wire resistance will go up with temperature.

 

[davenn]

The wire resistance will go up with temperature.

ahhh OK ... as I said a long time since I have played with that stuff

per chance there isnt a negative and a positive temp coeficient version ??

Dave

 

[Harald Kapp]

The "data sheet" says add heatsink for Iout = 3A.
However, the module has almost no way to dissipate the heat from the chip even at 2A or less. A well desigend PCB would show a large copper are under the back of the chip to drain the heat. I suggest you add a heat sink anyway, even if the current is well below 2A.
And measure the real current, don't trust the math - maybe the resistance is much smalller than 4 Ohm?

What does the 6V from the wall wart look like? Some (inexpensive) wall warts do have very bad or no regulation at all. Can you check for input ripple with an oscilloscope? The DC/DC converter expects a rather stable input voltage. Try adding a large electrolytic capacitor at the input (rule of thumb: 1000µF/A for a linear supply).

 

[davenn]

The "data sheet" says add heatsink for Iout = 3A.
However, the module has almost no way to dissipate the heat from teh chip even at 2A or less. A well desigend PCB would show a large copper are aunder theback of the chip to drain the heat. I suggest you add a heat sink anyway, even if the current is well below 2A.
And measure the real current, don't trust the math - maybe the resistance is much smalller than 4 Ohm?

yeah thats what i suggested .... wonder if he has yet ??

What does the 6V from the wall wart look like? Some (inexpensive) wall warts do have very bad or no regulation at all. Can you check for input ripple with an oscilloscope? The DC/DC converter expects a rather stable input voltage. Try adding a large electrolytic capacitor at the input (rule of thimb: 1000µF/A for a linear supply).

Ahhh but I didnt think of that

Dave

 

[duke37]

Resistance and temperature.
As I understand it, the electrons have trouble dodging the atoms when the atoms are dancing about due to being all hot and bothered so the resistance is higher

Wikipedia gives some details of the resistance of nichrome vs temperature. About 10% rise at 1000 deg C.

I think there can be magnetic effects on resistance. I forget how Constantan works.