Are there ARM controllers for small applications (like Cortex M0) available in small packages with maximum say 20 pins? I have the impression that in this area they don't quite are a threat for the usual suspects, like PIC and AVR.


Smaller packages, more specifically packages with less pins, are usually less expensive. Usually, because it also depends on the technology; QFP technology for instance is cheaper than CSP (Chip Scale Package). I presume this WLP (Wafer Level Package) for the LPC1102UK

WLP-16 package

is the smallest ARM package to date, body is 2.17 x 2.32 x 0.6 mm, with 16 bumps. That's damn small, yet it costs almost USD 5.00 quantity one (Digi-Key). Even at 3000 pieces the price is still over USD 2.00. (Remember, this is a Cortex M0, the lowest end ARM.)

From recent limited research I found that there are few Cortex M devices in very small packages, I haven't found anything like a SOT23-8, for example. Apart from the TI LM3S101 in a Fred Flintstone Package (aka SOIC-28) most packages seem to be QFP and QFN, and more of the former than the latter.
This is somewhat surprising, since the PCB assembly technology for both is the same, both can be inspected using flying probe, for example (which isn't possible on CSPs). Yet the QFN needs much less space than an equivalent QFP.

The explanation is demand, of course. Apparently most customers don't need the smaller space of the QFN (yet). Some manufacturers are pretty flexible about packaging, and may be prepared to introduce a new package for an existing device if you buy, say, 100k devices per year. This has more administrative than technical implications. So while ARM is widespread most customers will either need smaller quantities or don't really need the new package.
Still I expect ARMs to become available in smaller packages, like less than 20 pins. Especially for Cortex M0 this is going to be needed to successfully take 8-bitters the wind out of the sails. While SOT23 may not be an option I see many possibilities in QFN and particularly DFN.

DFN-10 package

Unlike DIL DFN is not limited to a specific width. This table

List of DFN packages

shows how many variants there are available from just 1 manufacturer. So there is always a solution for a specific number of pins and die size.
Small controllers like the LPC1102 would easily fit in a 3 x 3mm QFN-16, for instance, but apparently (and unfortunately?) this hasn't happened yet.

QFN-16 package

  • 1
    \$\begingroup\$ Fred Flintstone Package (aka SOIC-28) ....What? \$\endgroup\$ – Connor Wolf Jul 31 '11 at 3:35
  • 5
    \$\begingroup\$ @Fake - Well, it is a prehistoric package, especially for microcontrollers. The Fred Flintstone derives from the Fred Flintstone Format, or FFF, sometimes used to refer to the MM/DD/YYYY date format by ISO-8601 evangelists like myself. \$\endgroup\$ – stevenvh Jul 31 '11 at 6:09
  • \$\begingroup\$ @stevenh a small insight into semi cost. The cost of the chip in volume is usually 50% of the sale price. However in this case, depending on the process (I suspect 90nm) that chip costs NXP probably less than 60 cents. The margins especially on microcontrollers are enormous. However, say Apple were to buy that uC you mention with 16 pins, they would pay <1$. Now for the cost, the cost is usually 1/3 packaging (Csp lover due to a single RDL layer), 1/3 test (it takes sometimes multiple seconds to test these stuff) and 1/3 die. \$\endgroup\$ – Frank Jul 31 '11 at 6:35
  • 1
    \$\begingroup\$ @Frank - My experience with negotiating uCs goes only to 100k/yr, so Apple is a different league (100k for a small country like Belgium is huge, nobody ran that kind of production here. My phone was red hot from calls by suppliers :-)). For the same quantities my experience is that the same controller in a QFP48 package costs significantly less than the same die in a QPF64, so pin count matters (I expect the die to be the same). I also expect that an LPC1102 in a QFN16 package would cost less than the WLP16 we have now. \$\endgroup\$ – stevenvh Jul 31 '11 at 7:41
  • \$\begingroup\$ @stevenh I understand what you are saying. 100K is Belgium is good, but at some point of time, I was selling 5M/month to somebody in Finland :) Joke aside, I have provided the rudimentary analysis for people to understand cost dynamics of semi guys so that they can negotiate accordingly. \$\endgroup\$ – Frank Jul 31 '11 at 11:11

NXP LPC1102 16 pins http://www.nxp.com/documents/data_sheet/LPC1102.pdf

There are also several 32 pin M0 and M3 parts in NXP's range

However for very small apps, 8 bit MCUs often still have advantages, even if cost is similar, e.g. lower density packages, wider supply voltage, onboard eeprom, lower power consumption.

  • 1
    \$\begingroup\$ Lower power consumption in absolute numbers, yes, like the MSP430. In DMIPS/mW, I'm not sure.. \$\endgroup\$ – Federico Russo Jul 30 '11 at 10:56
  • \$\begingroup\$ The other advantage is being proven. I have made a mistake of designing something very new more than once and every single time, I end up debugging other people's problems. One big advantage of 8 bti, it has been there for several years and you know the software and hardware is fully vetted. \$\endgroup\$ – Frank Jul 31 '11 at 11:13

The LPC810 comes in a DIP8 package.


The smallest ARM microcontroller to date (March 2014) is the Freescale Kinetis KL03 micocontroller, based on the 32-bit ARM Cortex-M0+ core:

The Kinetis KL03 chip-scale package (CSP) MCU is the next world's smallest ARM Powered® MCU designed to support the latest innovation in smart, small devices. Available in the ultra-small 1.6 x 2.0 mm² wafer-level CSP, the Kinetis KL03 CSP (MKL03Z32CAF4R) reduces even more board space while integrating even more rich MCU features than previously seen in the market.

enter image description here


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.