Sample and hold module VREF

Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
Anonymous
Not applicable

HI,

   

I am checking on example project and found the sample and hold example. The sample and hold component can use internal and exteranl as Vref, but I could not find any reference on why and when you should use the internal or external vref. Any idea where can I find the information?

   

Tks

0 Likes
1 Solution
HeLi_263931
Level 8
Level 8
100 solutions authored 50 solutions authored 25 solutions authored

I know two reasons for using an external reference:

   
        
  • the internal one is not precise enough, ot to unstable (temp coefficent)
  •     
  • the internal one has the wrong voltage (e.g. you need 2.048V but the internal one only goes to 1.2V)
  •    
   

On the -050 kits there is even a spare footprint for an external reference with higher precision (and I think on the -030 kits too). The internal reference is accurate for about 0.9%, with a temp coefficent of 50ppm. The proposed LM4140 gives you 0.1% and 3ppm.

View solution in original post

0 Likes
17 Replies
HeLi_263931
Level 8
Level 8
100 solutions authored 50 solutions authored 25 solutions authored

I know two reasons for using an external reference:

   
        
  • the internal one is not precise enough, ot to unstable (temp coefficent)
  •     
  • the internal one has the wrong voltage (e.g. you need 2.048V but the internal one only goes to 1.2V)
  •    
   

On the -050 kits there is even a spare footprint for an external reference with higher precision (and I think on the -030 kits too). The internal reference is accurate for about 0.9%, with a temp coefficent of 50ppm. The proposed LM4140 gives you 0.1% and 3ppm.

0 Likes
ETRO_SSN583
Level 9
Level 9
250 likes received 100 sign-ins 5 likes given

1) You have to do an end to end error analysis to see what reference tolerance you need.

   

2) The internal reference can be scaled with PGA, its gain and drift would contribute to ref error analysis.

   

3) The S/H probably does not warrant need for high precision external reference, as its own errors

   

probably dominate end to end error analysis. But my conjecture can only be established by doing

   

the error analysis.

   

4) LM4140 A grade an expensive part, and only a 0 - 70 C part. Digikey shows ~ $4 - $5

   

part. The 10 ppm part more like $ 1.50 - $ 3.00.

   

 

   

The onchip reference is speced over I temp range, and its specs look like -

   

 

   

Anonymous
Not applicable

But how is the reference affecting the S&H?

   

May be I missed something here. Where is the reference come into play with the S&H process?

0 Likes
ETRO_SSN583
Level 9
Level 9
250 likes received 100 sign-ins 5 likes given

This should help -

   

 

   

Anonymous
Not applicable
        Thanks Bob. May be I should make it clearer at the beginning. I read the datasheet already. My question is. If I use internal reference . What is the effect on the S&H output, how is it being affected. Would it be +-x% of the input, where x is the error of VREF?. Would there be noise on the output which is the noise of the VREF? If I use external VREF of 2.5V, would it mean that the input must be 2.5V or above? Would it mean voltage below the VREF would be different to those ABOVE 2.5V? I got the small text box again....!!!!   
0 Likes
Anonymous
Not applicable

Also, is the effect of Vref post-samplig or pre-sampling?

0 Likes
HeLi_263931
Level 8
Level 8
100 solutions authored 50 solutions authored 25 solutions authored
        This took a while to find, it is hidden well in the TRM. When you look at the PSoC5 TRM, part 30.9 ("Sampled Mixer"), there is the statement: The use of the internal ground can cause different step sizes up versus down because the amplifier does not respond identically when the negative terminal jumps below ground. To avoid this distortion, use the external reference option and set it to 500 mV or greater. There is also a schematic showing how the S&H internally works (and I must to admit that I still not fully comprehend it). And sorry for the formatting, there is none right now for me. One should think hat after 4 weeks or so it should work. Can I redeem some of my reward points for a proper editor, please?   
HeLi_263931
Level 8
Level 8
100 solutions authored 50 solutions authored 25 solutions authored
        I should have used some quotation marks around the quote 😞 So what it comes down to (as far as I understand it): The internal reference for the S&H is not Vref, but ground (bad wording, I think, then). Using it seems to affect the linearity of the S&H, and can cause maybe jumping signals (the S&H samples at both the falling and rising edge, and the internal amplifier behaves not linear when the negative input goes below ground (which it will in the S&H configuration). One should guess this should be part of the component data sheet, and not be hidden in the TRM...   
HeLi_263931
Level 8
Level 8
100 solutions authored 50 solutions authored 25 solutions authored
        @summon dana: I'm pretty sure you know what this means exactly, you are way better in these things than I am...   
Anonymous
Not applicable
        Thanks hli, Shall check the TRM for details. As you said. it should be included in the data sheet or at least give a link to the TRM.   
0 Likes
lalec_300236
Level 1
Level 1
First like received

Went thru the TRM, it seems that using the internal one would be a bad option. I think it should state this in the datasheet and give more details of what would happen as well as the requirement of having external reference to more than 500mV.

Anonymous
Not applicable

I think Cypress should update the data sheet.

0 Likes
ETRO_SSN583
Level 9
Level 9
250 likes received 100 sign-ins 5 likes given

I filed a case with Cypress as there are two different schematics showing two different

   

T/H, S/H capacitor sizes.

   

 

   

I also address this step size issue. I think this reference is to mixer operation, CMR

   

allowable. But I have seen ap notes discussing an allowable CMR operation that

   

does not comply to datasheet.

   

 

   

Will keep forum informed when I get answer.

   

 

   

Regards, Dana.

ETRO_SSN583
Level 9
Level 9
250 likes received 100 sign-ins 5 likes given

I filed a CASE on this to clarify -

   

 

   

1) Two different schematics in TRM show two different cap sizes for S/H, T/H

   

 

   

2) I think the Vref comment about step size in mixer is a CMR issue of OpAmp,

   

not sure that it applies to S/H, T/H, just mixer. But I have been wrong before on guesses, 

   

so have asked Cypress to clarify. However the Vref issue in module datasheet

   

totally incomplete as to what it is doing when applied to S/H, T/H. I think it is a CMR

   

issue again with internal OpAmp.

   

 

   

Will keep forum informed.

   

 

   

Regards, Dana.

ETRO_SSN583
Level 9
Level 9
250 likes received 100 sign-ins 5 likes given

I thought the post had not occured, and lost the text, so retyped,

   

 

   

Sorry for confusion.

   

 

   

Dana.

Anonymous
Not applicable

 Thanks, dana.

0 Likes
ETRO_SSN583
Level 9
Level 9
250 likes received 100 sign-ins 5 likes given

.......

0 Likes