Types of Microcontrollers 1

[UsmanLula]

Hi all,

I have been trying to look for web pages which give information regarding the main types of microcontrollers that are available today i.e. Amtel, PICmicros etc and there advantages/disadvantages. Ive been sifting through many web pages all morning but somehow cant find that info!!!

Any help would be appreciated. Thanks in advance

Kind regard

Usman

 

[dbaird]

Check:

http://www.epanorama.net/links/microprocessor.html

David Baird
mrbaird@hotmail.com
Sr Controls Designer
EET degree.
Journeyman Electrician.

 

[cbarn24050]

Hi, your looking for the wrong answers. You need to match the controller to your application. That means you have to have an application first. If you just want to play around with 1 then buy a starter kit, the new zilog kit seems very good value.

 

[itsmoked]

UsmanLula... you are going to run into the problem of trying to find the corner to pee in, while standing in a sphere!

The neither the PIC's nor the Atmel are "better". Their truly astounding number of selections preclude a "better" company to go with. They both have better and less better offerings/price.

Your best bet is if you work with or associate with someone who uses one or the other then go with that one. This way you have some local experience to help you over the bumps you will surely find.

 

[UsmanLula]

Hi all

Thanks for your valuable comments and the web link. I finally found a web link which states the advantages and disadvantages of the common microcontroller types.

I have already developed a medical instrumentation using a PICmicro. However, I have to justify the use of PICmicro and why choose that and not something like Atmel micro etc. I have written a 35 pages report on the kit.

Is anyone experienced in PICmicros??? I may have a few technical questions on tolerance of the I/O ports (as i am running a few components off the PIC so that I can put the whole circuit to sleep when measurements are complete)....

Thanks once again for the comments. Much appreciated.

Usman

Medical Physicist

 

[MikeHalloran]

Justify? In 35 pages? Egad!

<justification>

Fight the enemy you know.

</justification>



Mike Halloran
Pembroke Pines, FL, USA

 

[UsmanLula]

Mikey

well not in 35 pages thats just a report on the work and results etc - but in my viva on 21st september!!! im gona be examined by consultant medical physicists - and il have to explain why i chose each component especially over other types in the market.

Ive already passed me Radiotherapy and Phys Meas portfolio/vivas. This is the final one. fingers crossed. wanna make sure i get it right - no resits!!!

Usman

 

[IRstuff]

Just a general comment. A 35-pg report suggests that you did not get the specific objections or concerns about the PIC vs ATMEL from whomever you're writing the report for.

So, how do you know that you've addressed the concerns of your customer?


TTFN

 

[UsmanLula]

IRstuff - good point

In our department, we only use PICmicros (especially in terms of experience). However, there must have been a reason for that! But in case i get asked a question in the viva about the choice of my microcontroller i can always say why a PICmicro is better (literaly be biased). lol

Also, I fulfilled the customer requirement using the PICmicro. The 35 page report is actually part of my medical electronics instrumentation 80 page portfolio. I did do two other projects involving a medical EMG amp and a Humidity sensor instrument (for breath study).

Anyway cheers bud.

Usman

 

[MacGyverS2000]

If everyone else in the department is using them, I would say that's a pretty good reason to continue using them, as well... assuming there's no overriding reason NOT to use them. In-house support obviously already exists, including both tools and experience... big items.

Beyond that, it's picking nits. One controller may have a better peripheral 'A', whereas the second may have a better peripheral 'B'. Sometimes the best choice is the one you don't have to make...


Dan
Owner

http://www.Hi-TecDesigns.com

 

[Warpspeed]

I agree, if you already have the programmer, documentation, and experience with the PIC that certainly gives you a head start.

Beginning completely cold with something totally different can be a very slow and frustrating learning experience.

I have been hopping back and forth between several different microcontroller brands recently on different projects, and it is quite disconcerting.

 

[itsmoked]

UsmanLula; I would just go with "PICs are as reliable and available as the Atmels" and our expertise is in PICs.
There is no discernable price difference. As your justifications.

 

[UsmanLula]

Thanks guys

Yes there are a few people with expertise in PICmicros in our department and we have all the tools for them too. One consultant specialises in them and works with them day in and day out.

What you guys say is quite true because I also had a time limit to complete the project and move onto the next one - also if i did run into problems with a new one microcontroller - i would be running around trying to find answers. However knowing that someone does have a lot of experience in PICmicros - I can just ask a collegue and find my answers. Its more productive too.

Many Thanks to Itsmoked, warpspeed, mcguyvers2000, and others.

Usman

 

[hrc]

ahhh..but...just found out today something new about PICs A/D.
In general, the I/O pins are very robust. We've even have put 9 to 12V on them and sure they get warm (ok..HOT) but they keep on tickin.

BUT..the analog pins are not as robust. They tend not to clamp transients very well, let alone over voltage on the pins and punch thru the substrate and cause the A/D reading to either be stuck at a value, and/or cause the A/D readings on the other channels to be offset.

how to prevent? Don't apply overvoltage to the A/D inputs. Otherwise, been very happy with PIC's for the last 15 years that I have been developing with them

 

[itsmoked]

That's interesting since the A/D pins have the same protection diodes.. Must have a lower impedance path to the A/D. Still I have never failed one in all my years of pin molesting. You must be a "super" pin molester.

I did once have a $3,500 National 4-bit emulator for a job I was doing. An underling borrowed the lab supply I was running it all with, without my knowledge. Then returned it set to 12V instead of 5V volts. He also hooked up everything so I wouldn't notice. When I turned it on... The emulator and I DID notice!

 

[richs]

Hiya to Smoked and hrc-

Just got through playing with the a/d on 16F877. The
data sheets mention that a MAX 2.5Kohm output impedance to
go into the analog input. That leads to some concern that
indeed the analog inputs are not the same critters as a
normal digital I/O pin.

For the little quick and dirty that I was doing, a
simple emitter follower did the trick.

And, as always it certianly helps to "Read The Fine Manual".
Microchip includes the info, you just have to remember
where you found it last time......

Cheers,

Rich S.

 

[itsmoked]

Hi richs

That 2.5k ohm provision is only for the leakage of the A/D circuitry. It has nothing to do with running digital loads on *those* pins.

If you have an analog load with more than 2.5k resistance then your readings may begin acquiring error.

 

[richs]

Yep-

I thought that was the topic of the last couple of posts.

The other postings was on the digital loading (I thought).

AND I stand corrected. The source impedance is spec'ed at
10K ohm, not 2.5K ohm. I did it backwards. I was
calculating a minimum charge time of the sample and hold
less than the example in the datasheets of 19.72 microsec.
The time requirements lent an input impedance of 2.5K ohm.

The devil is in the details.......


Cheers,

Rich S.

 

[itsmoked]

Hey richs
You can actually stretch the number a bunch. The main thing is the sample rate. If you have a high impedance source you can sit on the channel and charge the internal cap or you can feed a cap hung on the pin and that will essentially x out a large amount of your source impedance.