Convert double to byte array...

[Bonhomme]

Hi there,

I have a double precision floating point number that I want to convert into a byte array. Right now I'm under the impression that I have to convert the double to an integer first, then to byte array. The only problem is, my double is quite large (requires 48-bits) and there is no 48-bit integer in LabVIEW. But the main question is...is there a way to convert a double to a byte array? Thanks!

-Mike

[Al_S]

Use Advanced >> Data Manipulation >> Flatten to String
then Numeric >> Conversion >> String to Byte Array.

See the attached LV 6.1 example.

[Scott.Menjoulet]

Mike,

You can do this with one function by using the 'Type Cast' node (Advanced >> Data Manipulation). This is a little more flexible than the flatten function because you can cast directly to your desired type bypassing the string conversion. Also, you have more flexibility to choose your resulting type. You can convert to an array or cluster of numbers(integer or float) as well as strings. You can cast most basic LabVIEW types to/from most any other LabVIEW type. See the attached for an example.

Scott

[Bonhomme]

Thanks for the suggestion. However, I don't think I stated clearly what I needed. Say I have a double with a value of 10.00. When I use the data manipulation method, I get a byte array with values
40 (Hex)
24 (Hex)
BUT, what I wanted was a byte array with the value
A (Hex)
So in essence, convert a double to int, then to byte array. But to do that would require a 48-bit int. Any suggestions? Thanks.

[Bonhomme]

Thanks for the suggestion. However, I don't think I stated clearly what I needed. Say I have a double with a value of 10.00. When I use the data manipulation method, I get a byte array with values
40 (Hex)
24 (Hex)
BUT, what I wanted was a byte array with the value
A (Hex)
So in essence, convert a double to int, then to byte array. But to do that would require a 48-bit int. Any suggestions? Thanks.

[Al_S]

OK. To get a byte by byte raw binary representation of the double as if it were an integer, just keep dividing by 256. See the attached example.

[Greg_McKaskle2]

> So in essence, convert a double to int, then to byte array. But to do
> that would require a 48-bit int. Any suggestions? Thanks.

Try converting to a string, or use flatten to string, then convert the
string to a byte array. Because of endianness, you may need to reverse
the byte order depending on what you want.

Greg McKaskle

[Al_S]

Updated example which uses a while loop instead of a for loop so numbers > 48 bit are handled. Connector is wired so it can be used as a sub-VI.

[Bonhomme]

Thanks so much. It worked perfectly. I have another section where I have to do the same thing, but the integer is a 48-bit TWO'S COMPLEMENT number now. I can't just divide by 256 anymore...but have to worry about negation and stuff. Can LabVIEW help w/ anything like this? Thanks.

[Al_S]

Do you mean you want to take a negative number and represent it as 48 bit two's complement in a byte array?

Try this example. It's kinda messy but I think it works for positive or negative numbers.