gwideman
Posts: 45
Joined: Tue Mar 01, 2016 8:18 pm

Documentation bugs: RPi2&3 GPIO electrical specs

Tue Mar 01, 2016 10:10 pm

While attempting to find electrical specs for either the RPi3 I/O signals, or the GPIOs on the BCM2837 to which they are perhaps connected, I've noted the seeming complete absence of these basic specs.

On the basis that the RPi3 perhaps implements the I/Os like the previous model RPi2B or even earlier, I thought to look at those. But apparently there's no official data on the electrical I/Os on those either.

The only light being shed is by a very few unofficial docs regarding RPi2 I/Os. But there are serious problems.

http://www.scribd.com/doc/101830961/GPIO-Pads-Control2 posted in 2012 by Gert van Loo, describes the output voltage levels and drive current options for the GPIOs.

Puzzle: This doc excerpts the BCM2835 "full" data sheet, or possibly errata, but this and related electrical info isn't in any other doc that Google found for me. Just being able to get that basic level of I/O electrical data would be a help. (For those familiar with industry standard data sheets, that would be the Absolute Maximum Ratings and DC Electrical Characteristics sections. And there's no reason why the AC section shouldn't be included, plus typical I vs V plots and so on.)

BUG: Unfortunately, either that doc, or Gert's write up of it, is seriously mistaken, discussing VIL and VIH as though they are VOL and VOH. So that rather undermines the credibility of the data.

Then there's http://www.mosaic-industries.com/embedd ... ifications, a nicely written doc that, lacking BCM2835 docs, tries to divine what the I/O electrical specs might be by triangulating several non-BCM2835 info sources.

Puzzle: That doc comes to a range of different opinions on what VIL, VIH, VOL and VOH might be, and it's interesting that NONE of them correspond to Gert's excerpt from some maybe official top-secret Broadcom info.

FWIW, Gert's VIL, VOH look more like 3.3V version of "TTL" levels, whereas mosaic's look more like 1/3*VCC and 2/3*VCC "CMOS" levels.

So, herewith I'm reporting what I consider are bugs:

(a) RPi offers I/Os specifically for the purpose of connecting them to the outside world, yet without suitable standard electrical docs they are not fit for implementing that purpose in a disciplined and safe way.

(b) The unofficial docs (Gert's scribd and mosaic's docs) are either wrong, or conflicting to the point of having little or no credibility.

Have I missed some official docs somewhere that clears this up?

Graham

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 20741
Joined: Sat Jul 30, 2011 7:41 pm

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Wed Mar 02, 2016 10:08 am

What particularly are you missing? People have been bolting things to the GPIO's for the last four years with considerable success, so there must be enough information out there/

Have you tried the HAT docs?
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Please direct all questions to the forum, I do not do support via PM.

gwideman
Posts: 45
Joined: Tue Mar 01, 2016 8:18 pm

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Wed Mar 02, 2016 10:50 pm

Hi @jamesh,

Thanks for your response. As I said in my original post, I'm looking for: "Absolute Maximum Ratings and DC Electrical Characteristics sections. And there's no reason why the AC section shouldn't be included, plus typical I vs V plots and so on."

If these are unfamiliar, then download any datasheet from any chip family in the last 40+ years, and you will see what's included. As examples from diverse realms:

-- https://www.fairchildsemi.com/datasheet ... 4001BC.pdf pages 3-6

-- http://www.atmel.com/images/atmel-8271- ... mplete.pdf chapters 29-35 (for multiple variants)

These are the bare minimum specs always provided with electronic devices to facilitate designing their interoperation with other devices. In the absence of a schematic for a board which allows us to extract these specs from the component chip docs, the board itself needs such a characterization (which should be trivial for the board manufacturer to obtain from the chips).

I'm well aware that people have been, as you say, "bolting things to the GPIOs" for years. That demonstrates that with a few meaurements you can guess what will probably work sometimes. This provides little data as to where the limits are -- what happens across different manufacturing samples, at different temperatures, with different loads, at different speeds, and so on.

The basic specs I mention are the standard characterization that manufacturers always provide in order that their components or subsystems can actually be incorporated into a system by an engineer with assurance they will work in the variety of situations the engineer application entails. EEs do not proceed based on "hey, I got this one to work OK". We proceed by understanding the parameters that the components were designed to meet consistently, which allows us to create a design that complies with those parameters, and thus functions consistently.

This is such standard practice that it's extraordinary to me that there's any difficulty whatsoever finding such specs for the recent RPis.

On your suggestion, I did find "HAT docs" at https://github.com/raspberrypi/hats, but this mostly seems to concern I/O pin functions, and software control of I/Os. Important topics, to be sure, but I didn't see electrical specs -- perhaps I missed something?

User avatar
rpdom
Posts: 12861
Joined: Sun May 06, 2012 5:17 am
Location: Ankh-Morpork

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Thu Mar 03, 2016 4:59 am

gwideman wrote:Hperhaps I missed something?
The target market for the Pi systems? ;)

User avatar
karrika
Posts: 1046
Joined: Mon Oct 19, 2015 6:21 am
Location: Finland

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Thu Mar 03, 2016 6:00 am

paulv and joan have been trying to do this job by banging on the I/O pins. If you search posts by these user ID's you will probably get some idea of the electric characteristics of the pins. And certainly enough information to build external electronics outside the chip for reliable interfacing to the real world.

gwideman
Posts: 45
Joined: Tue Mar 01, 2016 8:18 pm

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Thu Mar 03, 2016 7:42 am

rpdom wrote:
gwideman wrote:Hperhaps I missed something?
The target market for the Pi systems? ;)
Hahaha, yeah... I can appreciate a bit of comedy in that. Except that selling RPi into the educational marketplace (let alone as a module to build products around), and then holding back the data required for electrical I/O calculations (and education about them) -- not so funny.

gwideman
Posts: 45
Joined: Tue Mar 01, 2016 8:18 pm

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Thu Mar 03, 2016 7:48 am

Thanks @karrika for the suggestion to hunt down threads with paulv and joan.

I ran across a number of recent interesting threads, including:

"Characterizing GPIO pins"
viewtopic.php?f=29&t=133740

... in which paulv and co make a start at trying to reverse-engineer the characteristics (which is helpful, but will not reveal the guaranteed levels and limits).

"Push Button Input Troubles"
viewtopic.php?f=28&t=131440&start=50

... in which paulv and co take about a month to figure out how to hook up a push button and get the debouncing working, due to lack of hard data.

These I would say are prime exhibits demonstrating why one would want basic electrical specs spelled out by the manufacturer, instead of users wasting their time trying to divine them (and still not succeeding).

The situation is very far from "certainly enough information to build external electronics outside the chip for reliable interfacing to the real world".

I now see this problem stretches a long way back, for example this thread from 2014:

Electrical specifications BCM2835
viewtopic.php?f=63&t=86721

... in which thexeno very reasonably asks for pretty much what I'm asking for... because it's what people accustomed to working with electronics understand is necessary.

Even Gert jumps in to say "When connecting devices there is no 'luck' involved. [...] If you want to connect the BCM28235 device to a product X you _have to look at the spec_ of both and see if they match."

Right. You look at the specs on both ends of the interconnection and design them to correspond. Trouble is, we don't have the RPi end of that.

Now, Gert _does_ say "All I/O of the BCM2835 are compliant with the standards. Thus all digital I/O will be compliant with the CMOS I/O standard for 3.3V CMOS devices." (at least, for RPi of 2014).

Trouble is, there's more than one possible 3.3V standard (as also noted by thexeno), and variations on those standards, and further the SoC has multiple options that can be programmed for both inputs and outputs.

Bottom line... the usual electrical specs, notably the limits, maxima and minima, are as necessary as ever.

User avatar
joan
Posts: 13578
Joined: Thu Jul 05, 2012 5:09 pm
Location: UK

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Thu Mar 03, 2016 10:19 am

I know nothing about the electrical specs being a software guy.

If what you need isn't already published then I guess you'll need to find an alternative platform. The situation in regard to the electrical specs doesn't seem to have changed since the Pi was launched.

Does the compute module documentation offer no insight? That is targeted at the embedded market.

gwideman
Posts: 45
Joined: Tue Mar 01, 2016 8:18 pm

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Sun Mar 06, 2016 5:16 am

Thanks @joan for your comment and suggestion to look at the RPi compute module for clues.

That module does at least have schematics, and a "Hardware Design Guide" (https://www.raspberrypi.org/documentati ... gnguide.md).

However, I still found no basic electrical specs for the GPIOs. All you can tell is that certain GPIO pins wire to the SoC. And there's no electrical specs for the SoC.

Graham

User avatar
Burngate
Posts: 5472
Joined: Thu Sep 29, 2011 4:34 pm
Location: Berkshire UK
Contact: Website

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Sun Mar 06, 2016 1:02 pm

http://www.scribd.com/doc/101830961/GPIO-Pads-Control2

If you need more, just like any piece of electronics, you'll have to investigate yourself.
However detailed, spec-sheets never cover all corner-cases.
If you think you might be designing yourself into a corner, you should redesign to aviod the corner rather than trying to find the parameters of the corner.

gwideman
Posts: 45
Joined: Tue Mar 01, 2016 8:18 pm

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Sun Mar 06, 2016 7:59 pm

Thanks @ Burngate for your reply.

While I'm sure there are some corner cases that spec sheets don't cover, they do a very good job of covering the main edge cases that are essential for interfacing to the components in question.

These specs are, at a minimum, VIH, VIL, and VOL, VOH and the currents at which those are specified. Also absolute max ratings for input and output voltages and currents. These are the limits that the manufacturer stays within, and that you can depend on when determining how to interface to the part.

We generally don't expect to have to "investigate for ourselves" the basic specs of a part required to apply it to the task for which it's advertised -- in the case of RPi, connecting it to other devices, and furthermore, learning about how to do that sensibly. The manufacturer's limits cannot be determined from measuring, unless you have a large and representative sample.

Perhaps didn't notice in my original post that I've already read the document you point to, and commented upon it.

That doc discusses the necessity of having VIH, VIL, and VOL, VOH (etc) at hand. For example "the max current at which the pad will still meet the spec". OK, what spec?

And then goes on to not only NOT present the specs, but actually discusses VIL and VIH (not to mention the third VIL is a typo) as though they are VOL and VOH. In short, not just inadequate but actually misinformation.

As an EE, I regard these specs as so essential, and so universally provided, that it boggles the mind that RPi doesn't provide them. But don't take my word for it, google for tutorials on how to interface to digital logic and you'll see they are referenced universally (if not always correctly!).

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 20741
Joined: Sat Jul 30, 2011 7:41 pm

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Mon Mar 07, 2016 9:48 am

This is being investigated.

Not an EE so still confused as to how we have managed 4 years of people interfacing to the GPIO's and this being the first request for these details though.
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Please direct all questions to the forum, I do not do support via PM.

gwideman
Posts: 45
Joined: Tue Mar 01, 2016 8:18 pm

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Mon Mar 07, 2016 10:28 am

jamesh wrote:This is being investigated.
Awesome, that will be a great service.
Not an EE so still confused as to how we have managed 4 years of people interfacing to the GPIO's and this being the first request for these details though.
Well, it's somewhat confusing how some people HAVE been trying to get this info for years, and it hasn't been forthcoming.

Many users just guess, or follow what works in other cases. It seems easy enough to get RPi to interface with 3.3V logic for example, or to drive a transistor that needs little current. What's harder is to assess where the limits are -- for example is there enough current (using what resistor value) for the base current of a transistor driving a large load. Can an RPi output interact with a 5V logic input (which family)? Or assessing the combinations of pull-ups and voltage requirements for a multi-device I2C bus that includes level shifters. Etc etc. Even in these cases you can get one or two RPis to work by hacking manually. That's different than specifying for production, or for a large audience, what values definitely will work without user fiddling.

People have been asking:
... and there are many more.

Even one of the two most-quoted reference (the mosaic site) is itself an attempt to assemble coherent electrical specs by looking at comparable SoCs (and ends up showing that comparable products are quite divergent, showing that it's not at all obvious what the RPis specs are).

Indeed, it's widely quoted that RPi GPIO VOH = 1.3V. This is highly unlikely to be correct, but raises considerable problems when trying to design for it. It may well be derived from Gert's error-riddled doc which mixes up the input parameters with the output parameters, and possibly VTH.

Example of confusion propagated: Mastering the Raspberry Pi

Bottom line: the present situation is a bit of a shambles. In the absence of real specs, rumor, hearsay and superstitious behavior prevail. :cry: - Graham

kaos
Posts: 108
Joined: Mon Mar 26, 2012 8:14 pm

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Mon Mar 07, 2016 12:43 pm

I would like to second this request, and am indeed glad that it is being looked into. Yes, we can, as witnessed by the horde of circuits published, get things working with the GPIO. But Graham is perfectly correct that it currently involves considerable guesswork and premature optimization for things that should be obvious from a glance at what should be the first resource for any electronics work; the datasheet.

--
Best regards, Kári.

hippy
Posts: 3898
Joined: Fri Sep 09, 2011 10:34 pm
Location: UK

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Mon Mar 07, 2016 4:33 pm

jamesh wrote:Not an EE so still confused as to how we have managed 4 years of people interfacing to the GPIO's and this being the first request for these details though.
Because most people don't need that information. That one gets a near 3V3 voltage out when high and near 0V when low, and putting near 3V3 and near 0V into an input gives a high and low reading is good enough for most people.

What a Pi puts out seems compatible with whatever digital electronics it is being fed into, and whatever digital electronics puts out seems compatible with what the Pi expects, and that's mostly what's needed to make interfacing work.

There may be edge cases but most people wouldn't usually be at those edge cases. I haven't seen any reports of people running into those edge cases but they might exist. In the case of 'black box electronics' which should be compatible - connect this to that and it will simply work - I haven't heard of cases where it hasn't.

Whilst I would like to see full specs on the GPIO electrical characteristics, consider them desirable to have, and otherwise limiting for those of us who would like to do some out of the ordinary things, they are not absolutely essential. If one were concerned one could simply buffer the signals in and out of a Pi. Not ideal but it is a solution.

rgarvin1
Posts: 2
Joined: Sat Mar 05, 2016 3:29 pm

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Mon Mar 07, 2016 5:41 pm

Graham,

Agree on the need for schematic and at least 'basic' electrical specs for the SOC device. MANY of my customers want me to use the RPi PCB as a basis for projects ( after all they say, it is very low cost and OPEN SOURCE). In reality it is NOT open source and without a schematic there is no way to properly evaluate its suitability for a project. For one offs and proto-types it is ok, but could not be used for projects in industrial control, as the quality and risk management evaluations would not be possible without a schematic.
I thought the whole basis of the RPi products was public availability of software source code AND hardware schematics.

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 20741
Joined: Sat Jul 30, 2011 7:41 pm

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Mon Mar 07, 2016 6:15 pm

rgarvin1 wrote:Graham,

Agree on the need for schematic and at least 'basic' electrical specs for the SOC device. MANY of my customers want me to use the RPi PCB as a basis for projects ( after all they say, it is very low cost and OPEN SOURCE). In reality it is NOT open source and without a schematic there is no way to properly evaluate its suitability for a project. For one offs and proto-types it is ok, but could not be used for projects in industrial control, as the quality and risk management evaluations would not be possible without a schematic.
I thought the whole basis of the RPi products was public availability of software source code AND hardware schematics.
The Pi has never been an open source project. There is a binary blob to run the GPU (and actually now another one to drive the Wifi chip pn Pi3), and the schematics are not yet available (although might be later on). The whole idea is to be a great platform for education., not to be open source, although that would be a nice to have.

Pi have been used extensively in industrial control, despite the lack of 'openness', although a lot of projects remain under the radar. 5 Ninja's Slice device is a available example, which uses the compute module.
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Please direct all questions to the forum, I do not do support via PM.

gwideman
Posts: 45
Joined: Tue Mar 01, 2016 8:18 pm

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Mon Mar 07, 2016 7:43 pm

I for one don't care if the RPi hardware is open source. What I care about is whether the exposed surface of the product is documented enough to use for its advertised purposes. Those purposes include interfacing to "the world", and teaching about such interfacing. Proceeding intelligently based on elementary principles with either of those activities is blocked for lack of specs (and actively undermined by the wrong specs that are out there and widely referenced).

I hope that the need for such specs isn't discounted based on the commenters saying they haven't heard of people needing them. There are a lot of medical treatments I haven't heard of either, but I suspect there are many people who need them :-). Unsurprisingly, a lot of people haven't seen elementary EE, and proceed by trial and error. Needless to say, that's not a solid basis for deploying RPi in serious applications, nor for teaching.

So again, I strongly encourage jamesh in his offer to look into this matter. Thanks.

User avatar
PeterO
Posts: 4257
Joined: Sun Jul 22, 2012 4:14 pm

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Mon Mar 07, 2016 8:11 pm

gwideman wrote:I for one don't care if the RPi hardware is open source. What I care about is whether the exposed surface of the product is documented enough to use for its advertised purposes. Those purposes include interfacing to "the world", and teaching about such interfacing. Proceeding intelligently based on elementary principles with either of those activities is blocked for lack of specs (and actively undermined by the wrong specs that are out there and widely referenced).
If you think you need highly detailed exact specs then you need to learn how to design for real world situations where devices come with a range of characteristics. Good design is all about making your circuits work when you don't know the exact values or parameters of the components.

PeterO
Discoverer of the PI2 XENON DEATH FLASH!
Interests: C,Python,PIC,Electronics,Ham Radio (G0DZB),Aeromodelling,1960s British Computers.
"The primary requirement (as we've always seen in your examples) is that the code is readable. " Dougie Lawson

User avatar
davidcoton
Posts: 3194
Joined: Mon Sep 01, 2014 2:37 pm
Location: Cambridge, UK

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Mon Mar 07, 2016 9:14 pm

PeterO wrote: If you think you need highly detailed exact specs then you need to learn how to design for real world situations where devices come with a range of characteristics. Good design is all about making your circuits work when you don't know the exact values or parameters of the components.

PeterO
The specs that gwideman is asking for are the bits that define the range of characteristics. It's not a big problem for one-off or low volume hobby projects, but a properly designed medium or high volume commercial product will be designed to work across that range, and across the range of specs for the other half of the interface, so that any random combination of parts that meet their respective specs will work together. If the range of any component is not properly characterised by the spec, such design is not possible, and mass production becomes a gamble. It may not be a big cgmble, but there is no way to quantify the risk in advance. A mistake can be enough to break a company.

Part of EE education is to teach/learn the proper design techniques, using the interface specs. If these are not available for the Pi, then the Pi cannot be used to teach that aspect of design. That is where the information is important to the Foundation's goals. It will incidentally increase Pi sales because the risk in using the Pi in commercial products is reduced -- RPF may or may not consider that a benefit.
"Thanks for saving my life." See https://www.raspberrypi.org/forums/viewtopic.php?p=1327656#p1327656
“Raspberry Pi is a trademark of the Raspberry Pi Foundation”

User avatar
PeterO
Posts: 4257
Joined: Sun Jul 22, 2012 4:14 pm

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Mon Mar 07, 2016 9:23 pm

Would you expect the GPIO pins to be anything out of the normal ?
This isn't an issue for the Foundation.
Don't you think Broadcom would have considered these issues when designing the SoC ?
Making something with a non-standard and difficult to interface signals would kill their sales !
PeterO
Discoverer of the PI2 XENON DEATH FLASH!
Interests: C,Python,PIC,Electronics,Ham Radio (G0DZB),Aeromodelling,1960s British Computers.
"The primary requirement (as we've always seen in your examples) is that the code is readable. " Dougie Lawson

User avatar
davidcoton
Posts: 3194
Joined: Mon Sep 01, 2014 2:37 pm
Location: Cambridge, UK

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Mon Mar 07, 2016 10:48 pm

PeterO wrote:Would you expect the GPIO pins to be anything out of the normal ?
Define "normal". That's what specs do. Without the right spec, which one would you use?
PeterO wrote:This isn't an issue for the Foundation.
Don't you think Broadcom would have considered these issues when designing the SoC ?
The RPi is an RPF product. Yes, the spec will come straight from the Broadcom spec, for direct I/O with no other onboard components. But Pi users are not Broadcom customers, and don't have Broadcom specs. Making the specs available probably requires agreement between RPF and Broadcom.
PeterO wrote: Making something with a non-standard and difficult to interface signals would kill their sales !
PeterO
No-one is suggesting it is non-standard. But I say again, WHICH standard?
"Thanks for saving my life." See https://www.raspberrypi.org/forums/viewtopic.php?p=1327656#p1327656
“Raspberry Pi is a trademark of the Raspberry Pi Foundation”

gwideman
Posts: 45
Joined: Tue Mar 01, 2016 8:18 pm

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Tue Mar 08, 2016 12:42 am

@PeterO:

The responses by @davidcoton address your main comments very well.
PeterO wrote:If you think you need highly detailed exact specs then you need to learn how to design for real world situations. [...] Would you expect the GPIO pins to be anything out of the normal?
Comments such as these suggest that you are unfamiliar with VIH, VIL, VOH, VOL and other basic (not "highly detailed exact") specs that capture the real world variation that you rightly advocate designing for. (In addition to just documenting the basic voltage thresholds and currents to be expected.)

You likely still "get things working" if you only plug in same-family logic, guess luckily for other components, or stick to what others have shown usually works. But you won't know why it works, or why it doesn't when it fails outright (or worse, intermittently), and how close your circuit is to failing.

So a nice outcome of this conversation would be for you (and others in the same boat) to become familiar with these specs so you could get closer to performing design tasks. Which would be greatly aided if RPi actually published such specs.

By way of introduction to the ubiquity of these specs, I challenge you to find a single chip (or other system for that matter) that you've ever hooked up to an RPi that doesn't have these specs available.

User avatar
PeterO
Posts: 4257
Joined: Sun Jul 22, 2012 4:14 pm

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Tue Mar 08, 2016 7:24 am

gwideman wrote:
PeterO wrote:If you think you need highly detailed exact specs then you need to learn how to design for real world situations. [...] Would you expect the GPIO pins to be anything out of the normal?
Comments such as these suggest that you are unfamiliar with VIH, VIL, VOH, VOL and other basic (not "highly detailed exact") specs that capture the real world variation that you rightly advocate designing for.
My BSc is in Computing and Communications Engineering, so your suggestion is wrong. Maybe everything was still TTL 5V logic levels when I got it ;) but the principles are still the same today.
All I will say is that if you have never had to work with incomplete specs. then you are very lucky.
Unfortunately in these days of "intellectual property" being seen as more important than releasing useful engineering data you will need to learn how to do a certain amount of reverse engineering for yourself.

PeterO
Discoverer of the PI2 XENON DEATH FLASH!
Interests: C,Python,PIC,Electronics,Ham Radio (G0DZB),Aeromodelling,1960s British Computers.
"The primary requirement (as we've always seen in your examples) is that the code is readable. " Dougie Lawson

steaky1212
Posts: 16
Joined: Wed Sep 07, 2011 5:18 pm
Contact: Website

Re: Documentation bugs: RPi2&3 GPIO electrical specs

Tue Mar 08, 2016 8:17 am

PeterO wrote: Maybe everything was still TTL 5V logic levels when I got it ;) but the principles are still the same today.
All I will say is that if you have never had to work with incomplete specs. then you are very lucky.
PeterO
The issue is that we don't know if the logic levels are TTL or CMOS - it makes a difference. And unfortunately you can't base an entire design because one works on your bench. That one might have especially poor (or good) characteristics.

Return to “Advanced users”