Teensy 4.0 external ADC and voltage protection

Status
Not open for further replies.

ghostintranslation

Well-known member
Hi,

I'm doing a board that will accept a range of 0-5v as analog input. As Teensy 4.0 accepts only 0-3.3v I was looking at ways of making this work. So I was looking at linear optocouplers, opamps, resistor dividers... And finally I want to use an external I2C ADC which work with this range. It's the MCP3426A0T-E/MS which is a differential ADC.

Additionally I want to make sure no more than 5v and no less than 0v is getting in so I'm using 2 diodes in front of the ADC (let say I try to input 9v or -9v by mistake). I think the maximum the ADC can withstand is 5.5v which should be what it get if I try 9v before the diodes right?

I would like to confirm if this is going to work before I go buy components and PCBs?

External ADC.jpg

On this schema, MUIN1C and MUIN2c are getting the voltage range, and the 5v is coming from Teensy 4.0 .

Thanks
 
Its preferable to use schottky diodes, they have a much lower forward voltage and will conduct before the internal
anti-static protection diodes. 10k in series is possibly adding some noise, I'd go for more like 1k sort of range.

(Having said that 10k will likely allow the internal diodes to cope without overload, which is a smaller part count).

The BAT54S is often used for signal protection as it has two schottkys in series (SMT only though).

BTW your circuit seems to be lacking the required decoupling for the chip. And you are using the supply
as a differenrial reference voltage - so you'll be limited by its stability and accuracy.
 
It's a simple circuit. Put it into LTSpice and try it.

I didn't know LTSpice thanks, I made a simulation that seems to work as expected.


Its preferable to use schottky diodes, they have a much lower forward voltage and will conduct before the internal
anti-static protection diodes. 10k in series is possibly adding some noise, I'd go for more like 1k sort of range.

(Having said that 10k will likely allow the internal diodes to cope without overload, which is a smaller part count).

The BAT54S is often used for signal protection as it has two schottkys in series (SMT only though).

BTW your circuit seems to be lacking the required decoupling for the chip. And you are using the supply
as a differenrial reference voltage - so you'll be limited by its stability and accuracy.


Schottky diodes seems more interesting indeed and the BAT54S looks nice, I like that there is the 2 diodes in one package, I'm looking at saving space.

What do you mean by "decoupling for the chip"?
 
All logic chips and most analog chips require supply decoupling for stable functioning. ADCs and DACs in particular require careful attention
to this and the datasheet should explain whats needed.
 
All logic chips and most analog chips require supply decoupling for stable functioning. ADCs and DACs in particular require careful attention
to this and the datasheet should explain whats needed. Values, types and proximity of the various capacitors are all important if you want
to get the performance listed in the datasheet.

See figures 6.1 onwards in the datasheet: https://www.mouser.co.uk/datasheet/2/268/22226a-81911.pdf
 
All logic chips and most analog chips require supply decoupling for stable functioning. ADCs and DACs in particular require careful attention
to this and the datasheet should explain whats needed. Values, types and proximity of the various capacitors are all important if you want
to get the performance listed in the datasheet.

See figures 6.1 onwards in the datasheet: https://www.mouser.co.uk/datasheet/2/268/22226a-81911.pdf

Ah I see, and also I only notice now that this chip has an internal voltage reference of 2.048V which Vin should not exceed... This chip might not be the right choice for me in the end...
 
Well actually I could do a voltage divider with 2 resistors like in figure 6.5 , it's just that it means 2 more resistors to add and I'm very tight on space (even SMT) .

I see also the ADS1115IDGSR is interesting, similar chip but its internal voltage reference is changing according to VDD, so with VDD=5v Vref = 6.144v and Vin can go from 0 to 5v, so no external voltage divider required in this case.

There are so many ADC it's hard to decide on one.
 
OK so finally I don't know why I had excluded doing a voltage divider, but in the end it works just fine for what I'm doing... I had to take a break to see it...
So I'm just lowering the input before the diodes so that 5v is reduced to 3.3v, and the diodes cap it to 3.3v in case of trying to input something out of the 0-3.3v range before going in Teensy, I tested it on LTSpice and then tried it for real and it's all good.

This thread has been helpful to discover LTSpice and also for the choice of diodes (BAT54S), and on the side understanding better the differential ADCs.

Thank you
 
Status
Not open for further replies.
Back
Top