How the way to identify a logic device tcd time in its datasheet?

Thread Starter

raziell122

Joined Mar 28, 2023
71
Hello everybody,
how the way to find the contamination delay time (tcd) of a logic device in its datasheet?
I am looking for it for logic gates, Flip-Flops, ROM/RAM etc..
I need it for a final project in logic-design course and I can choose any datasheet and any manufacturer for each component and it doesn't have to be a specific model or company.
My lecturer said its sometimes called "TPLZ", but I can't find it on every datasheet, also sometimes "TPLZ" appears on a timing diagram but its value does not mentioned on the sheet.. on the other hand ChatGPT says its called "TPLH" so I am very confused now..

Thanks in advance!
 

ronsimpson

Joined Oct 7, 2019
3,199
I do not understand " contamination delay ".
Here is a data sheet for an IC. The delay through the part is clearly in the data sheet.
1714012766707.png
 

WBahn

Joined Mar 31, 2012
30,286
Hello everybody,
how the way to find the contamination delay time (tcd) of a logic device in its datasheet?
I am looking for it for logic gates, Flip-Flops, ROM/RAM etc..
I need it for a final project in logic-design course and I can choose any datasheet and any manufacturer for each component and it doesn't have to be a specific model or company.
My lecturer said its sometimes called "TPLZ", but I can't find it on every datasheet, also sometimes "TPLZ" appears on a timing diagram but its value does not mentioned on the sheet.. on the other hand ChatGPT says its called "TPLH" so I am very confused now..

Thanks in advance!
I don't think I've ever heard of "contamination delay". If you describe/define what it means, we might be able to determine what it is normally called in our part of the planet.

TPLZ is, probably, the propagation delay between an input going LO and the output going high-impedance.

Don't rely on ChatGPT to ever provide accurate information.

You say that you can't find it on every data sheet. But, you said you could choose ANY data sheet, so choose one that has it.

Also, provide a link to such a sheet here so that we can look and see if we can help you better understand it, as opposed to having to pull out our crystal balls.
 

WBahn

Joined Mar 31, 2012
30,286
Have you tried Googling it (instead of asking ChatGPT)?

I searched for 'contamination delay' and the very top hit was:

https://en.wikipedia.org/wiki/Contamination_delay

I've done quite a bit of what it is talking about when examining circuits for race conditions, I've just never heard it called by that term.

Google (or any other search engine) is a MUCH better friend than ChatGPT or any of its brethren.
 

Thread Starter

raziell122

Joined Mar 28, 2023
71
Have you tried Googling it (instead of asking ChatGPT)?

I searched for 'contamination delay' and the very top hit was:

https://en.wikipedia.org/wiki/Contamination_delay

I've done quite a bit of what it is talking about when examining circuits for race conditions, I've just never heard it called by that term.

Google (or any other search engine) is a MUCH better friend than ChatGPT or any of its brethren.
Yes, that's exactly the term that you attached from wiki, do you maybe know what is a common term that they usually use in datasheets? or if it has the same meaning as "TPLZ"?
Sorry for my poor English..
 

WBahn

Joined Mar 31, 2012
30,286
Yes, that's exactly the term that you attached from wiki, do you maybe know what is a common term that they usually use in datasheets? or if it has the same meaning as "TPLZ"?
Sorry for my poor English..
Again, post a link to a data sheet that you say has TPLZ on the diagram, and we can see if it is consistent with the definition.
 

WBahn

Joined Mar 31, 2012
30,286
This data sheet for example, it has a timing diagram with TPLZ on page 5
First, note that those figures are generic and some of them may or may not be applicable to a given part (read Note H).

As I speculated earlier, TPLZ is the time delay between when an output that is being driven LO is placed in a high-Z state (i.e., disabled). IF the part has an output enable, this parameter is known as t_dis in the data sheet (read Note E).

These are NOT the same as the contamination time (though they would need to be considered if you are using parts that have high-Z states and you have race conditions that need to be mitigated).

The contamination time is the MINIMUM time between when an input changes and when the output reflects that change.

So, for the part in this data sheet, look for the entry that tells you the MINIMUM amount of time between when an input changes and the output changes.

Note that the reason that a lot of folks don't know the term "contamination delay" is because we were educated back when we were expected to understand concepts instead of memorize formulas, so because we understood what a race conditions was, we could determine which parameters in the data sheet were relevant to the specific race condition we were evaluating. Today, texts are more apt to invent new terms and then throw out formulas that use them, instead of requiring students to be able to come up with their own formulas based on the problem they are solving. This isn't a slam on you -- it's just an observation to offer an explanation for why many people don't recognize that term.
 

Thread Starter

raziell122

Joined Mar 28, 2023
71
First, note that those figures are generic and some of them may or may not be applicable to a given part (read Note H).

As I speculated earlier, TPLZ is the time delay between when an output that is being driven LO is placed in a high-Z state (i.e., disabled). IF the part has an output enable, this parameter is known as t_dis in the data sheet (read Note E).

These are NOT the same as the contamination time (though they would need to be considered if you are using parts that have high-Z states and you have race conditions that need to be mitigated).

The contamination time is the MINIMUM time between when an input changes and when the output reflects that change.

So, for the part in this data sheet, look for the entry that tells you the MINIMUM amount of time between when an input changes and the output changes.

Note that the reason that a lot of folks don't know the term "contamination delay" is because we were educated back when we were expected to understand concepts instead of memorize formulas, so because we understood what a race conditions was, we could determine which parameters in the data sheet were relevant to the specific race condition we were evaluating. Today, texts are more apt to invent new terms and then throw out formulas that use them, instead of requiring students to be able to come up with their own formulas based on the problem they are solving. This isn't a slam on you -- it's just an observation to offer an explanation for why many people don't recognize that term.
Thanks for your reply.
It does not surprise me that my lecturer was wrong when he said that Tcd = Tplz.
About the part, what should I do if I can't find anything about minimum amount of time between when an input changes and the output changes?
 

WBahn

Joined Mar 31, 2012
30,286
About the part, what should I do if I can't find anything about minimum amount of time between when an input changes and the output changes?
Look harder. It's there.

Look at the waveform figures. Find a parameter that is the time from when an input changes (which uses the midpoint voltage as a reference), to when the output changes (which also uses the midpoint voltage as a reference).

Or, look at the tables and find a parameter that is for the time it takes the output to change in response to a change in the input and then look at what the minimum time is for that to happen.
 

Thread Starter

raziell122

Joined Mar 28, 2023
71
Look harder. It's there.

Look at the waveform figures. Find a parameter that is the time from when an input changes (which uses the midpoint voltage as a reference), to when the output changes (which also uses the midpoint voltage as a reference).

Or, look at the tables and find a parameter that is for the time it takes the output to change in response to a change in the input and then look at what the minimum time is for that to happen.
From the waveform figures I can see that the parameters that indicate the time from when an input changes to when the output changes are TPLH or TPHL, but it is a bit weird for me, because we were taught that that these two parameters usually refer to propagation delay time (tpd) and not to contamination delay time tcd. Also in the table on page 4 I can find information only about tpd minimum value.
Am I wrong?
 

WBahn

Joined Mar 31, 2012
30,286
From the waveform figures I can see that the parameters that indicate the time from when an input changes to when the output changes are TPLH or TPHL, but it is a bit weird for me, because we were taught that that these two parameters usually refer to propagation delay time (tpd) and not to contamination delay time tcd. Also in the table on page 4 I can find information only about tpd minimum value.
Am I wrong?
Read Note G. The notes are there for a reason.

What is the definition of the contamination delay?

What is the definition of the minimum propagation delay?
 

Thread Starter

raziell122

Joined Mar 28, 2023
71
Read Note G. The notes are there for a reason.

What is the definition of the contamination delay?

What is the definition of the minimum propagation delay?
Yea now I see that the note says that TPHL & TPLH are the same as tpd.
In our course, the definition of contamination delay is the minimum time that passed from the moment the input of the component left a stable value until the output of the component left a stable value, but my lecturer never mentioned the "minimum propagation delay", does it have the same meaning as tcd?
TBH we have never used datasheets on this course, they always gave us tpd & tcd in the specific exercise.
 

WBahn

Joined Mar 31, 2012
30,286
Yea now I see that the note says that TPHL & TPLH are the same as tpd.
In our course, the definition of contamination delay is the minimum time that passed from the moment the input of the component left a stable value until the output of the component left a stable value, but my lecturer never mentioned the "minimum propagation delay", does it have the same meaning as tcd?
TBH we have never used datasheets on this course, they always gave us tpd & tcd in the specific exercise.
Again, look at the meaning of both things. Don't worry about what they are called, look at what they mean. Are they the same or aren't they? If you don't think they are the same, what is it that you think is different between them?
 

Thread Starter

raziell122

Joined Mar 28, 2023
71
Attaching 2 graphs from my lecture notes:
for Tpd:
TPD.png
Our definition is Tpd >= Max{d1,d2}

for Tcd:
TCD.png
Our definition is Tcd <= Min{D1*, D2*)

The difference that I find between them is that Tpd relates to the time passed between Vin has finished to change (reached to a stable logic value) until Vout has finished to change, whereas Tcd
relates to the time passed between Vin has started to change (left a stable logic value) until Vout has started to change.
I would maybe think that Minimum Tpd in the datasheet relates to the best performance of the propagation delay time of the specific device, of course you are probably right and they are the same (Tcd = minimum Tpd), but its not so easy for me to catch the reason :)
 

nsaspook

Joined Aug 27, 2009
13,536
Have you tried Googling it (instead of asking ChatGPT)?

I searched for 'contamination delay' and the very top hit was:

https://en.wikipedia.org/wiki/Contamination_delay

I've done quite a bit of what it is talking about when examining circuits for race conditions, I've just never heard it called by that term.

Google (or any other search engine) is a MUCH better friend than ChatGPT or any of its brethren.
I've never seen it called by that term either. The contamination I see talked about here seems to be a delay that's influenced (and used in circuit design) mainly by external circuit design inducted issues instead what's inside the IC package. It's not something I would expect to see on a device data-sheet.
 

Attachments

WBahn

Joined Mar 31, 2012
30,286
Attaching 2 graphs from my lecture notes:
for Tpd:
View attachment 320700
Our definition is Tpd >= Max{d1,d2}

for Tcd:
View attachment 320701
Our definition is Tcd <= Min{D1*, D2*)

The difference that I find between them is that Tpd relates to the time passed between Vin has finished to change (reached to a stable logic value) until Vout has finished to change, whereas Tcd
relates to the time passed between Vin has started to change (left a stable logic value) until Vout has started to change.
I would maybe think that Minimum Tpd in the datasheet relates to the best performance of the propagation delay time of the specific device, of course you are probably right and they are the same (Tcd = minimum Tpd), but its not so easy for me to catch the reason :)
I think the fine point that you are getting tripped up on is related to subtle distinctions between essentially the same definition, specifically, what constitutes "starting to change".

From the Wikipedia article:

In digital circuits, the contamination delay (denoted as tcd) is the minimum amount of time from when an input changes until any output starts to change its value. This change in value does not imply that the value has reached a stable condition. The contamination delay only specifies that the output rises (or falls) to 50% of the voltage level for a logic high.
{/QUOTE]

Notice that the criterion here for "starts to change") is 50% of a logic high.

Now look at the data sheet waveforms and you will see that they use 50% (for 5 V devices, and pretty close to 50% for 3.3 V devices) for their propagation delay specifications.

From a nice academic view point, you could get nitpicky and specify the start and stop times as being related to the various guaranteed logic thresholds, but in the real world this would be highly unlikely to be done. It would require a LOT more measurements and take up a lot more room in the data sheet for almost no actual utility because the differences are very minor and no decent logic designer is going to push the limits, so any rational margin of safety in the design is going to swamp the differences.
 
Top