IEEE 11073 32bit float format

[kiran.holla]

How to convert 32bit hex number to IEEE 11073 32bit float format?

[tyk007]

---

@kiran.holla wrote:

How to convert 32bit hex number to IEEE 11073 32bit float format?

---

I think you mean IEEE 754.

Can you show us what you have tried thus far?

[kiran.holla]

Thanks tyk007,

No, I ment IEEE 11073 32bit float format.

I found the way, actually in IEEE 11073 ( 1st byte is Exponential (10 pow 1st byte) and rest 3 bytes are mantissa).

[GerdW]

Hi kiran,

please check for yourself: Wikipedia IEEE 11073

When you want to convert to a base10 32bit float you need to find a library which suports this conversion (use Google or any other search engine) - or use some bit-banging to program this conversion on your own…

When you already created a VI for this conversion: please attach it here for future reference!

[crossrulz]

---

@kiran.holla wrote:

I found the way, actually in IEEE 11073 ( 1st byte is Exponential (10 pow 1st byte) and rest 3 bytes are mantissa).

---

No sign bit?

[kiran.holla]

Thanks GerdW for the link.

I have attached what fitts to my solution. There could be number of way.

The attached VI requires Hex array passed, will give you Temperature in Celsius. I believe even signed value is taken care.

Note: 4 Byte Array(Hex) would be in BigEndien IEEE 11073 . i.e  value of 24.25 would be 79 09 00 FE

Let me know if there is any problem in it.

[GerdW]

Hi Kiran,

can you attach your VI for LV2014?

How are Base10 floats related to IEEE11073? Wikipedia doesn't mention them (when reading the article rather quickly)!

[PhillipBrooks]

Stackoverflow has a thread regarding the conversion of 11073 to a java float - 

http://stackoverflow.com/questions/11564270/how-to-convert-ieee-11073-16-bit-sfloat-to-simple-float-...

[jamiva]

Your code is not 100% correct. It works fine for positive numbers. BUT the mantissa is a 24bit signed integer. You are converting the mantissa string to a U32 integer and ignoring the sign bit in the string. This c-code came from the "STACKOVERFLOW" thread that PhillipBrooks mentioned. It is the conversion for a 12 bit integer. However, you can easily change the constants for a 24 bit conversion.

[GerdW · Accepted solution]

Hi Kiran,

Let me know if there is any problem in it.

Well, I found a problem of using too much unneeded datatype conversions!

Is there a reason you need to convert U8 values to strings to be able to build other integer and DBL values from them?

Why not something simple like this:

The other case holds your version of this code…

(The shown code can be simplified even further, I just replaced your U8-to-string-to-U32 dance…)

Remark: both code version give the same result.

None give the result you expect according to your own statement:

Note: 4 Byte Array(Hex) would be in BigEndien IEEE 11073 . i.e  value of 24.25 would be 79 09 00 FE

So I guess there are other problems too in your VI…