Microchip buys Atmel

[MrAl]

Hello again,

More generally, perhaps Microchip should look at the success of Atmel in the hobby market and go with that, taking it up to the professional level. "Marduinos" that can be used in commercial equipment and appliances. They'd make a friggin fortune because manufacturers could lower their development time by orders of magnitude.

Any other name ideas?
Mc Arduino's (har har)

 

[Nigel Goodwin]

We have to know the current capability of each individual pin and we have to know the current capability of the whole chip, it cant be any other way because it's too hard to make a chip that can handle 500ma total. So i see this not as a lie but as a matter of reading all the spec's not just one

Presumably the 'complaint' is that it's NOT in the datasheet, that only the individual pin rating is?.

Certainly PIC datasheets give both - I've never looked at an AVR one, but always presumed it was similar?.

 

[be80be]

This is kind of funny but just what if microchip stops selling the 8 bit atmel chips why would they want too there 8 bit chips are cheaper and have lots more made into them.
Uarts pwm stuff like that. And then make a new 32 bit wifi chip using the best of both.
I guess the arduino boys would have to start paying up a little more to keep there toys.
I've played with both I have a 18f2550 that runs the same code line for line as uno and I don't see it any harder to use then a uno..

 

[MrAl]

Presumably the 'complaint' is that it's NOT in the datasheet, that only the individual pin rating is?.

Certainly PIC datasheets give both - I've never looked at an AVR one, but always presumed it was similar?.

Hi,

Here is what i find on the data sheet verbatim:

02/2013
ATmega48A/PA/88A/PA/168A/PA/328/P
30.1 Absolute Maximum Ratings*
30.2 DC Characteristics
Operating Temperature . . . . . . . . . . . -55°C to +125°C
Storage Temperature . . . . . . . . . . . . . -65°C to +150°C
Voltage on any Pin except RESET
with respect to Ground. . . . . . . . . . -0.5V to Vcc+0.5V
Voltage on RESET with respect to Ground -0.5V to +13.0V
Maximum Operating Voltage . . . . . . . . . . . . . . . . 6.0V
DC Current per I/O Pin. . . . . . . . . . . . . . . . . . . 40.0mA <------------------<<<
DC Current Vcc and GND Pins . . . . . . . . . . . 200.0mA <------------------<<<

I'll add that these max conditions can not be sustained.

Try finding the max clamp diode rating though.

 

[be80be]

That's good but 8 leds @ 40 mA is what and most people using a arduino think its ok LOL
Just in case they miss the math it's 120 mA over the rating of the chips max

DC Current Vcc and GND Pins . . . . . . . . . . . 200.0mA <------------------<<<

 

[Ian Rogers]

This is kind of funny but just what if microchip stops selling the 8 bit atmel chips why would they want too there 8 bit chips are cheaper and have lots more made into them.

I should imagine that Atmel will continue as is for some time... MicroChip will have customers to keep they won't want to loose any!! We won't see much change for a bit!

 

[NorthGuy]

The DIY people seems to be of two kinds - some use very old PICs even though there are newer better models (perhaps because of old Internet tutorials), while others move towards high end (Ethernet, Graphics etc.) and then often move away from PICs to ARM-based controllers which seems to have more hype now. The core of Microchip products are not used much for DIY. I'm sure Microchip is losing their share on DIY market, and Atmel is in much better positioned for DIY. But how big the DIY market can possible be anyway - these are all small purchases. Is it worth spending 3.5 billion for that?

 

[Andrei_D12]

The DIY people seems to be of two kinds - some use very old PICs even though there are newer better models (perhaps because of old Internet tutorials), while others move towards high end (Ethernet, Graphics etc.) and then often move away from PICs to ARM-based controllers which seems to have more hype now.

The DIY people you are talking about (those who use old PICs such as 16F84) are used to using them. In my opinion, they have mastered that microcontrollers, so creating a project with them is a lot easier than using something newer, just like in the case of a regular driver who meets the latest generation of cars.

As for the development boards, the PIC based ones are pretty expensive compared to the Arduinos. When it comes to their capabilities, I have absolutely no idea yet, because I am learning to program 8bit controllers (learn riding the bike, then the car... up to the spaceship) .

 

[be80be]

It's not the do it yourself market they are after It like I said you take a cheap pic less then a $1 and add atmel and you can come up with a one cheap WiFi chip and that's where there going to make big money.

It's in the writing on the wall They want that loT Powerhouse they now have the pieces they need to make that happen. I can't wait to see a wifi chip that works out of the box and you can read it's data sheet LOL.

 

[NorthGuy]

It's in the writing on the wall They want that loT Powerhouse they now have the pieces they need to make that happen. I can't wait to see a wifi chip that works out of the box and you can read it's data sheet LOL.

They already had all the things to create WiFi chips and IoT stuff. And they already sell WiFi modules.

 

[be80be]

They already had all the things to create WiFi chips and IoT stuff. And they already sell WiFi modules.

Sure they did that's not the point they now have lot better options to make faster and better one chip solutions.
Ok let's look at what they had wifi chips that cost $11 to $28 a pop.
Then you have stuff like esp-12 that's $2.50 a pop.
Now they have the same playing ground they could and I bet they will make chips that are better then anything out there.
This is what they have now 32 bit arm chips
then you have

World-class embedded security solutions

• Atmel CryptoAuthentication™ — Offers product designers an extremely cost-effective, easy to design, tiny, and ultra-secure hardware authentication capability.
• Trusted Platform Module — The Atmel Trusted Platform Module (TPM) provides strong hardware-based public key (RSA) security on a single device for personal and tablet computers as well as embedded processor based systems.
• Atmel CryptoRF — Atmel CryptoRF® is a 13.56MHz RFID device family employing a 64-bit embedded hardware encryption engine, mutual authentication, and up to 64Kbits of user memory.
• Atmel CryptoMemory — The Atmel CryptoMemory® family offers a range of cost-efficient, high-security electrically erasable programmable read-only memory chips (EEPROMs) and host-side security for applications requiring comprehensive data protection.

You can now nock the rest out of the water.

 

[MrAl]

That's good but 8 leds @ 40 mA is what and most people using a arduino think its ok LOL
Just in case they miss the math it's 120 mA over the rating of the chips max

Hi,

Well, some authors just oversimplify their articles if you want to call them that. This will happen for any technology.
But if you are suggesting that sometimes this is done on purpose, with Atmel i would not doubt it. Microchip seems to know how to write data sheets and articles better than Atmel anyway, so at least we may see a change for the better there. Then again the take over of National Semi by TI did not produce better data sheets on the old parts yet as far as i know, except for maybe a few. Then again then again (that's ''then again' squared) National made better data sheets than TI all along so why mess it up right...he he.

Since the ARM chip has *no* current rating for the two input clamp diodes, i'll have to take them as being rated for the pin current because the individual pin current is given, usually around 3ma. That's my only guess at this point. Microchip CLEARLY shows this spec on their data sheets, which allowed me to have more respect for Microchip than for Atmel. Microchip may be able to create a separate spec for this, but i dont know if they will even bother or if they will have time to get to it in the near future.

In any case this is a big change for the uC industry. How that is going to affect the products and data sheets we wont know for a while yet. First thing though MC will probably get right down to branding all the data sheets with their own name (chuckle).

Another point is that Microchip would not have bought Atmel if they did not LIKE their chips. That says something positive about the future of the Atmel chips. I would doubt that they would just want to buy to put them out of business...too big of a customer base to loose.

 

[Tony Stewart]

Just like in the case of PIC where the max curent per output is 25mA , but, if all outputs were sourcing/sinking 25mA each, the controller would turn into a candle.

AVR's use ALVC2 CMOS technology which at low voltage gives an RdsON of 25 Ohms rather than ALVxx of 50 Ohms, thus it is true each driver can handle 40mA with an 800mV rise and 32mW per driver. Total power dissipation must factor how many drivers are ON simultaneously drawing this power.

To verify yourself, read the Vol/Iol specs for worst case near the end of the spec.

 

[Andrei_D12]

AVR's use ALVC2 CMOS technology which at low voltage gives an RdsON of 25 Ohms rather than ALVxx of 50 Ohms, thus it is true each driver can handle 40mA with an 800mV rise and 32mW per driver. Total power dissipation must factor how many drivers are ON simultaneously drawing this power.

To verify yourself, read the Vol/Iol specs for worst case near the end of the spec.

When playing with controllers, I use LEDs in order to check that the micro is giving a feedback. I am not planning to drive motors, actuators or other components with higher current consumption than a LED directly from Mcu. Microcontroller just controls the action of a circuit. Controllers have a higher output current than other ICs.

 

[Nigel Goodwin]

The DIY people you are talking about (those who use old PICs such as 16F84) are used to using them. In my opinion, they have mastered that microcontrollers, so creating a project with them is a lot easier than using something newer, just like in the case of a regular driver who meets the latest generation of cars.

That doesn't really apply - you simply move to a more modern 14 bit PIC, and it's pretty well identical, just with more facilities - which was my reason for using the 16F628 in my tutorials, long ago

 

[Andrei_D12]

Oh, sorry. That's true.
As for the output curent, I don't see the point in having a larger curent per output, when the modern LED curent consumption is very small and it is very bright. Perhaps for driving more leds without a buffer transistor...

 

[Tony Stewart]

Oh, sorry. That's true.
As for the output curent, I don't see the point in having a larger curent per output, when the modern LED curent consumption is very small and it is very bright. Perhaps for driving more leds without a buffer transistor...

Andrei
The point of ALVC logic design used in Atmel chips is to offer higher speed logic with lower power loss at low Vcc (<3.3V) with lower ESR or RdsOn values in the Nch,& Pch CMOS FETs !! But you wanted to drive one port with an LED at more than the rated current that provides avalid "logic " level, you could. You may not beenfit from understanding this principle, but perhaps others might. WHen you drive a load with more drop than standard logic levels but is not a logic device , rather an analog device and do not exceed the absolute worst case current based on ampacity of tiny gold or aluminum filament wirebonds , that works OK.

It is not to simply drive LEDs at 5Vcc.

The 16F628 PIC specs are pasted below,

Using the linear RdsON characteristics and Ohm's Law, one can calculate the worst case ESR or RdsON for a "1" , VOH and a "0", VOL
If VOL = 0.6 V at IOL = 8.5 mA, VDD = 4.5 V, -40° to +85°C then ESR=70Ω
- remember this is worst case but the nominal value is closer to 50Ω at 5.0V, 25°C )

This contributes to the voltage drop with the load and thus determines a more accurate expected load current, as well as proves why Microchip 5V logic cannot drive as much current or at a lower characteristic impedance as Atmel at 3V for high speed logic when matched transmission track impedances are used. This is not exclusive to Atmel but rather, extra low voltage CMOS technology in controlling the CMOS drivers with lower RdsON values. The lower voltage logic "family called 74ALVCxxx is <25Ω over temp at 3.0V and rises in ~110Ω as Vcc drops to 1.65V

FYI, with modern high efficacy 5mm LEDs , used as indicators, you only need 2~3mA, whereas 20mA is painfully bright.

 

[MrAl]

Hello again,

After reading Tony's posts i remembered that a lot of chips have different ratings for each pin but with one overall rating for the whole chip which is lower than the sum of all the pin ratings. The reason this seems strange at first is because we usually use more dedicated chips where the one pin might be the output and a collector and rated for 500ma, and that's it. Another pin, 200ma, and that's it.
The microcontroller option is there to allow more versatility to the designer, who has the option of use either pin for say 30ma and then others at only maybe 5ma. If it wasnt like that then we'd have to divide 200 by 14 and only get 15ma max output per pin, which puts more of a limit on the design application. We can do 15ma per pin ourselves, so they chose not to limit us like that and go with the higher rating so at least some of the pins can be used for driving something harder. If that wasnt the case, we might have to try to get away with using two 15ma pins to get higher outputs.
It's all part of the thermal management which comes with all designs. In this case they gave us more versatility but we still have to use our heads when it comes to the total power management.

Another example would be an inside peripheral that say did RS232. One block inside the uC chip, but available on any of 12 external pins. We could choose to use that one RS232 port on say pin 3, pin 4, pin 5, etc., but we cant use it on all those pins simultaneously because there is only one block inside that does an RS232 port.

The temperature of the die inside is related to the power dissipation and the surface area of the chip on all sides and top and bottom. If we are stuck with a given surface area, given power dissipation per pin, and maximum die temperature, then we are stuck with a given total power dissipation. If we went beyond that, the temperature would rise too high and burn up the chip. Yes a fan would help, but that's usually not what we want to do.

 

[Tony Stewart]

The substrates silicon wafers for CMOS are usually P type doped silicon for improved insulation, but when the input or outputs of CMOS exceed the supply threshold of 1 silicon diode their is an SCR Thysristor switch effect with a PNPN vertical layer junction which effectively is a shoot-thru failure between Vcc and Vdd. Since CMOS has been made upto today, they use Schottky diodes on all inputs to each rail to shunt this and inhibit this failure mode often caused by ESD or any source of voltage > 0.5 V outside the supply rail. JUst like in voltatile memory logic and Thrystors alike this memory effect is cleared when supply is removed and so the failure mode stops and normal operation resumes.... unless the metallic layers and semiconductive doping layers are fabracated to give lower ESR, then the net ESR is low enough to cause the chip to exceed 150'C and the epoxy to start to smell, or even get hotter and start to crack.

We always used mixed vendors in HCxxx CMOS during the 70's& 80's but knew if we wanted fastest performance to use certain vendors like Fairchild becuase they had lower ESR , slightly more current drive but also slightly more self-destructive when an SCR latchup incident took place. One such time was when we were field testing a SCADA , ISDN broadband WAN with AMR, teleshopping and graphical weather data and a high speed serial port to the TV for the class of computers in those days that we had such events. Static on the TV and discharge to the remote serial keyboard to the cards in the basement which experienced SCR latchup. The ones like Motorola CMOS which had 300 Ohms per port at 12V would not fail but get warm, and others like Fairchild with less than 150 Ohms per port ESR would get too hot and self-destruct. We were able to fix the ESD problem with better ground wire to the basement to the TV channel changer serial port and keyboard for teleshopping.

But it wasn't until years later that I understood why some were more likely to be non-destructive than others. Vcc^2/ESR= Pd and 1Watt into a chip rated for 250mW was going to rise 4x as much. This all has to do with the design of the chip and its doping concetration levels and the effective RdsOn of each switch and the inherent defective SCR latchup effect from over voltage. Not only can you get latchup on an input stage , but also an output stage ( but harder ) YOu can also toggle the state of a Flip Flop by back drive an ESD pulse into Q from ESD, which is why we must use an extra inverter to buffer Flip Flops driving long wires to prevent this defect in design, inherent in all CMOS logic for Flip Flops.

Even those the planar junctions are depletion mode FETS, the vertical layers when biased outside the supply range form the SCR latchup effect with bipolar characteristics.

Now today for bipolar transistors if we want exceptionally low saturation resistance in a driver , we use FETs but if the switching capacitance is too high and we need high voltage, we turn to Transistors made by Xetex (or now Diodes Inc) super-saturated devices with Rce values in milliohms so with 1 Amp it is now possible for Vce to be <20mV. Although different than RdsOn for a FET, the patents are similar in describing the doping and channel topology which makes their devices better but also much more expensive but low Vce/Ice and high Vce ratings with very low capacitance

 

[Tony Stewart]

I recall going to U of T EE grad student thesis day about 10 years ago and the biggest funders of R&D and free tools was Atmel and Microchip. THis hopefully led graduates to design with their products and result in more sales.

It also spawned the Microchip’s PIC vs. Atmel’s AVR wars when under-graduates had to make a choice on which technology and tools they were going to learn.
WHere Atmel had more performance bang and user friendly open source tools, Microchip had more bang for the buck. When it comes to high volume, every penny counts in the decision which choice of technology to use. Now ARM and MIPS just sell their license to use their architecture to the respective companies, is there a chance the merger can avoid the royalty costs with a new architecture?

I once had an Audi 5000 that had something like 21 microcontrollers in it. I think Microchip would like to steal marketshare in automotive and smartphone applications which have been dominated by an unfamiliar name here called Runesasu Erekutoronikusu Kabushiki Gaisha, who were the world's largest semiconductor producer for cars a couple years ago. It's easier to just say Renesas Electronics as in "renaissance advanced solutions" formed by a collaboration effort of NEC, Hitachi and Mitsubishi.

So Atmel has to get smarter and cheaper to compete in future high volume products but they wont expect sales to Nissan or Toyota as they are investors in **broken link removed**

Isn't this cute **broken link removed**

Trivia..
the typical ratio of Flash to RAM is 8:1 for most solutions.