DS18B20 - how to test authentic?

[Yellofriend]

I just got a DS18B20 sensor probe from Digikey, and the manufacturer is DSRobot, a China based company. I admit, my fault, I didn't looked them up.

So my question is, is there a simple test to check if the part is authentic or not?

[Red Squirrel]

Those are normally made by Maxxim.  I think they were made by Dallas before. (think Dallas got bought out? I forget)  If you bought from Digikey then you should have gotten a legit product.  If you suspect it's bad maybe contact them so they can check their supply.   Never heard of DSRobot.

[Nusa]

Dallas Semiconductor was purchased by Maxim in 2001. So DS and MAX parts are all made by Maxim now.

[Yellofriend]

Also, why DS18B20? There are cheaper ...

I didn't buy the TOS  part, it's a wired probe in a stainless steel housing. I need it waterproof (actually yogurt proof). So there are a bit less options for me, and at $6.50 or so it doesn't matter too much. I also buy parts commercially from Digikey so I have never issues with minimum orders.

In the meantime I found this blog with a test procedure, will try it later today:

https://blog.peku33.net/aliexpress-ebay-china-market-fake-ds18b20-sensors/

[janekm]

Also, why DS18B20? There are cheaper ...

I didn't buy the TOS  part, it's a wired probe in a stainless steel housing. I need it waterproof (actually yogurt proof). So there are a bit less options for me, and at $6.50 or so it doesn't matter too much. I also buy parts commercially from Digikey so I have never issues with minimum orders.

In the meantime I found this blog with a test procedure, will try it later today:

https://blog.peku33.net/aliexpress-ebay-china-market-fake-ds18b20-sensors/

I doubt DSRobot would sell a completely non-working one (as that blog seems to test for). It's more likely to get a relabelled lower-accuracy sensor from the same range. You can read out the family code from the device ROM (check the data sheet) to check for that.

[raptor1956]

Also, why DS18B20? There are cheaper MAX31820 and MAX31820PAR out there which are virtually identical to DS18B20 and DS18B20PAR.

I was not aware of the other parts and just did a little checking ... it seems the other parts are not spec'd quite as good as the DS18B20.  The DS18B20 is spec'd at +/- 0.5C from -10C to +85C whereas the other parts are +/- 0.5C from 10C to 45C.

I am doing some Peltier module testing using five of the DS18B20's as I type this...


Brian

[Yellofriend]

Last night I did the multimeter diode test and compare the DS18B20 probe with some TOS DS18B20 that I got a few weeks ago from Digikey. In that test the probe behaves the same as the lose Digikey parts. So the probe might be using an authentic DS18B20.

[janekm]

Last night I did the multimeter diode test and compare the DS18B20 probe with some TOS DS18B20 that I got a few weeks ago from Digikey. In that test the probe behaves the same as the lose Digikey parts. So the probe might be using an authentic DS18B20.

That test only tells you it's a device that behaves similar to one with a one-wire interface. Could be DS18B20 or something else, such as MAX31820. At least read out the family code, though I suppose it's possible there are actual clones out there which will return the same family code...

[rs20]

You could look into rise/fall times, see exactly when it reads the value of the one-wire bus by scanning through different pulse periods, etc. Measure supply current as well. Look at radiated noise frequencies on the powerline. Nothing will ever provide perfect proof one way or another, but measuring lots of these weird properties will provide plenty of data points that you can then check whether they fall "within the bell" of a selection of genuine parts.