Using 5V TTL with I/O and Logic Analizer

[PicoPete]

I am considering purchase of the Analog Discovery II but have some questions:  

I am interested in using the Logic Analizer with conventional TTL logic outputs (and inputs) to be read by the Analog Discovery II; The specs for the Analog Discovery II shows  Input logic: LVCMOS (1.8V/3.3V, 5V tolerant); so my questions are:

1 - Is it safe to presume that the "5V tolerant" indication means I can input a logic level from the output of a TTL logic chip directly without an intermediate level translator ?  

2 - Can the 3.3v output (which does not indicate "5 Volt Tolerant" in the specs) directly drive a single TTL gate, or do I need to insert a voltage translator on the output to interface with a conventional TTL or TTL-LS, etc. type gate ? 

3 - if for some reason the Analog Discovery does not suit my requirements Is there any return privilege?

Please advise,

Thank you

[JColvin]

Hello,

I have moved your question to the Scopes and Instruments section of the Forum where the engineer best suited to answer your question will be able to see and respond to your question more easily.

As for your question 3, we will work with you through an RMA as needed.

Thank you,
JColvin

[PicoPete]

Hello JColvin:   I would think that my question would be part of an existing knowledge-base, but unfortunately it's not.  I would have to presume, too, that Digilent knows the answer to this really basic question which is not complicated; either the logic analyzer inputs/outputs will work with standard 5V TTL or they will not.  And I apologize if my reply seems somewhat pushy as I do realize the engineers are busy people, too.  However, I can move on to some other product that works, too, but before I do so is there any way to get an answer today?  I don't want to place an order blindly without knowing the answer to my question and receive Analog Discover II then determine that it does not work with 5V TTL and resulting in my having to return it and be hit with a 15% restocking fee.  Would appreciate any expedite to my question.  Thanks again.

[Bianca]

@PicoPete,

1. It is safe to use the Logic Analyzer with TTL logic outputs. The only thing to mention is that 5V will open the protection diodes and will take about 8mA from the driver. It shouldn't be a problem though. You can check the Chapter 5 in the Reference manual here.

2. The FPGA knows only 3.3V. If your TTL receptor has the minimum and maximum threshold to 3.3V, you can use it, if it doesn't you can't.  

3. Regarding the return privilege, you can ask for a refund in 14 days with a 15% restocking fee, like you said above 

Best regards,

Bianca

 

[D@n]

@PicoPete,

"I would have to presume, too, that Digilent knows the answer to this really basic question ..."  Not necessarily.  Most engineers pick and choose from parts that come off of a shelf.  (Think Digikey, but there are many more shelves than just those Digikey provides)  Electronic components today declare which standards they work with, such as the 5V TTL or 3.3V CMOS standards.  The parts simply declare what standard(s) they work with, not how well they work outside of their advertised performance range.  Whether or not these parts and pieces work with standards to which they were not designed is not a normal question that gets answered when bringing a product to market.  The question that normally gets answered is, does this product work within the standards for which it was designed?  To presume that a device works outside of its advertised regime is to presume that a test bench was built to test it outside of this regime.  Think that through, though, every test bench costs money, the more test benches you build the more your product costs.  In reality, there are just too many possibilities to test against, and so products are only sold as conforming to a certain standard.  To assume anything more is being presumptuous.

The devices and components, however, that you are looking at are declared to be "5V tolerant".  In general, this means either that i) the components and parts used were declared to be 5V tolerant by their respective manufacturers, or ii) that a protection circuit has been introduced so as to keep the part from being damaged with a 5V input.  It does not mean that measurement devices will properly measure voltages that are outside their limits.

So, to come back to your question, the unofficial answer to your question is always yes and no.  (I'm not Digilent, I cannot offer official answers ...)

1. Logic works based upon the middle voltage between true and false.  Hence, 2.5V is the threshold for 5V logic and 1.65V is the threshold for 3.3V logic.  From this standpoint, 3.3V logic outputs *should* generate an active high for 5V logic inputs, since 3.3V is greater than 2.5V.  Hence the answer is that 3.3V outputs "should" be able to drive 5V gates.
2. 5V Binary inputs to "5V tolerant" parts *should* also work.  The thresholds are similar enough, etc.
3. Now, with all that said, I've worked on projects in the past where voltage translation issues that *should* work ... didn't.  Wires and other on-board components have resistances associated with them that will adjust voltages, comparators tend not to be at exactly the threshold they specify, etc.  Even if the Digilent board produces accurate an accurate 3.3V output, the board you want to connect it to may have a problem accepting a 5V input.
4. As an example, on one project I worked on years ago a 3V input was applied to a 1.8V FPGA bank, with the startling response that the excess voltage leaked out of the other pins in the FPGA causing other logic to fail.  This sent our whole team trying to figure out why the part of the circuit, not connected with the "broken" wire, wasn't working.  We spent weeks looking for the problem in the wrong place.  Sure, the engineer had placed a resistor between the 3V output and 1.8V input that was supposed to convert between the two ...

Given your harsh words above, I can only imagine there's some poor Digilent employee trying to put together a hastily engineered test on his bench today to try to come up with a quick answer for you.  Give the guy a break, and know the final answer:

While it will work in theory, your actual mileage (without the voltage translator) may vary.  The only way to truly know the answer is to place the parts and pieces on your bench and try them.  Like I mentioned above, though, because Digilent doesn't necessarily have the components you will be connecting together, they really cannot give you an authoritative answer--although I imagine they will try.

Dan

 

[PicoPete]

Hello Dan:  Thanks for your long opinion:  

First, Logic Analizers are used for various logic families such as current 3.3v CMOS as well as others including the somewhat older, though still relevant TTL families.  

Second, while the Analog Discovery II (and I) are NOT "dedicated" Logic Analizers, they state that they perform logic analysis (as well as Protocol analysis) so it's perfectly within the scope of the indicated feature to wonder how compatible the particualar feature is when dealing with non-3.3v CMOS, especially in view that standard TTL is still alive and relevant.  

Third, the specs indicated "5V tolerant": "tolerant" can mean different things depending on what the "tolerant" was designed to be tolerant with and in what way. My original question was specifically related to this particular feature as it may apply to 5V TTL logic levels.  

Forth, I'm sure the engineers at Digilent, presuming they designed the Analog Discovery II (and I) are aware that 5V TTL is still a viable logic family and that there's lots of 5V TTL-based logic in the real world and that it's not going away anytime soon.  Now, whether they actually tested the Analog Discovery II (or I) logic analyzer using 5V TTL is an unknown of course, but since they mentioned 5V tolerant, as a potential purchaser I have the right to inquire to what extent that "tolerant" means, hence my question.  

Also, when one is looking to purchase pay money for a piece of test gear, confirming whether or not a particular feature is suitable and asking for a quick answer is not harsh but to the point - after all time is money.  

Relegating a technical pre-sales question/answer process to a forum to get a definitive answer (vs someones opinion which is not usually relevant)  can be time consuming when one is looking for an answer that will decide whether to make a purchase or not hat could result in having to return a product.  

Providing a user manual can sometimes help in finding an answer but not always.  

Equipment vendors have engineers, often with dedicated telephone numbers, that can answer a technical pre-sales question.  Of course on a relatively inexpensive item such as the Analog DIscovery II such support would be costly, hence support goes to a forum which is OK.  

But let's put it another way and say YOU would like to purchase a piece of test equipment for a non-specific application, the gear may or may not work for you, the answer can not really be determined from the user manual, specs, or application notes if provided and you need to know ASAP to meet a deadline, so your only recourse is:

1) Purchase the item and test it yourself with the possibility that you have to return it and pay a restocking fee if it turns out it's really not going to do what you hoped it would but more importantly that you could have avoided the return shipping cost, restocking fee, and time had you asked the vendor up-front, or

2) Based on the answer from the vendor which is mostly positive, make the purchase, or if it's negative look elsewhere for a different product - plain and simple.  

Digilent did, in fact, provide an answer that made perfect sense based on their knowledge of their product circuit topology - something that the prospective  purchaser does not have privy to or can determine from documentation sometimes.  

As a result of Digilent's answer, a mostly affirmative one, I will purchase the Analog Discovery II.  

So I think characterizing my words as  "Given your harsh words above..."  is not relevant (nor constructive);  My words weren't meant to be harsh and their tone is not harsh. 

Now if you have direct experience with using the Analog Discovery II with 5V TTL and the logic analizer I'd be interested in hearing about it and I'm sure others would be too.

Enjoy     

 

[Henrik Olsen]

@ D@n: The thresholt on 5V TTL is NOT just 2.5V, please see attached image.

@PicoPete: My experience (with AD 1) is that 3.3V on the digital output lines are NOT sufficient to drive a High Level Input on 5V TTL.
According to specs eg the 74AHC00 requires at least 3.85 V (driven on 5.5V Vcc)

When I get hold on an AD 2, I'll try to adjust and lower Vcc output on AD 2 from 5V til 3.3V to drive the TTL logic. That could solve the problem.

Henrik

 

[D@n]

@Henrik Olsen,

And here I just thought the threshold must be somewhere between the 0V and 5V and naturally assumed it would be halfway in between ...

So, at your instigation,since I'd never ever looked this up before, and because your source didn't cite a "5V TTL Standard" reference of any type but rather one manufacturer's implementation of TTL thresholds, I thought I might do a little digging via Google.  It seems there is some confusion about what the threshold is.  For example,

TTL gates operate on a nominal power supply voltage of 5 volts, +/- 0.25 volts. Ideally, a TTL “high” signal would be 5.00 volts exactly, and a TTL “low” signal 0.00 volts exactly. However, real TTL gate circuits cannot output such perfect voltage levels, and are designed to accept “high” and “low” signals deviating substantially from these ideal values. “Acceptable” input signal voltages range from 0 volts to 0.8 volts for a “low” logic state, and 2 volts to 5 volts for a “high” logic state.  [Found in All About Circuits]

and this source suggests only that the threshold be somewhere between 2V and 3.3V.

So ... I guess my conclusion would be something like, "Your mileage might vary."

Dan