My first ARM/LPC design, critic requested.

[paulie]

Coltan is used to make pinhead capacitors

I wonder if that is a reference to their size or to the engineers who choose to incorporate them into designs.

[nctnico]

Tantalum are blood capacitors
http://news.bbc.co.uk/2/hi/africa/1468772.stm

Not only that but in order to mine Coltan the people cut forrests which provide shelter for many endangered species.

[Kjelt]

AFAIK tantalums have their bad reputation from their old predecessors that had a tendency to explode and have limited life.
The present generation tantalums are pretty good AFAIK and I see them often in quality A brand products, so wonder why they would do that?

Tantalums will still explode when mounted in reverse. And yes that happens even in an automated process. Another problem with tantalums is that you need to control the current slew rate into them. If the slew rate is too high they might go off. A tantalum is basically a thermite lance/bomb because they consist of a fuel and an oxidizer. If they go off they can burn right through a PCB making it necessary to replace the entire PCB.

Besides that MLCC is cheaper nowadays and has a much lower ESR. So why bother with tantalums. The only reason to use a tantalum is when you need a relatively large capacitance for which there is no MLCC alternative. Another reason could be that you need a slightly higher ESR due to a regulator not being stable with MLCC capacitors. Still the strong influence of voltage on capacitance makes it tricky to select the proper MLCC. The capacitance is usually highly overrated.

Bottom line: I never use tantalums in my designs unless there is no alternative.

you repeated almost my entire posting.

[nctnico]

But added good reasons why they should be obsoleted.

[Kjelt]

So to say it all in one post and concerning the solid smd tantalums:
http://www.act-source.com/news/tantalum-capacitors

Tantalum capacitor advantages and disadvantages
 
Advantages:
 » Volumetric efficiency: offering a much high level of volumetric efficiency than many other types and being superior to common electrolytic capacitors.
 » Good frequency characteristics: they are more suitable for use in a number of applications where electrolytics could not be used.
 » High reliability: Tantalum capacitors are able to provide an almost unlimited life being operated within their ranges. Therefore their use is not time limited as in the case of the electrolytic capacitor.
 » Wide operating temperature range: often specified for operating over the range -55C to +125C. This makes them an ideal choice for use in equipment for use in harsh environmental conditions.
 » Compatibility with modern production methods: Tantalum capacitors are able to withstand the temperatures of SMT production and are there fore ideal for use in many new electronics designs.
 
Disadvantages
 » Low ripple current ratings: tantalum capacitors do not have high ripple current ratings and should not be used in areas that require any levels of current to be passed.
 » Not tolerant to reverse or excess voltage: Tantalum capacitors don’t like excess voltage or reverse, even spikes can destroy them. If they are exposed to excess or reverse voltages then they easily can explode.
 » Higher costs: more expensive than many other capacitors.

[paulie]

If you replace the aluminum electrolytic in that comparison with monolithic chip then nearly all the advantages become disadvantages. Monolithic are better than tantalum in the same way those had it over electrolytic. At least electrolytics as mentioned are still king of the mountain when it comes to large capacitance. Not many 1,000uf or 10,000uf chip caps.

[nctnico]

And with proper selection and carefull placement you can design long lasting equipment with electrolytic capacitors in them.

[paulie]

True. True. And of course the same can be said of thermionic valves.

One of the nice things about SMD if that for a board like this you don't have to decide what type of cap to use until you actually build it. With a little care pads can be placed that will accept ceramic or tantalum chip caps or even SMD electrolytics. Just make sure the minimum spacing is small enough to handle the tiny ones.

[neslekkim]

So what should one do in hobbyist design?, or small designs?, use smd electrolytic?, those are in those can design right?, quite bit bigger than the tantalums?, or are there others one should use instead?
(Mainly thinking about when you try to reuse other designs, and you don't have enough knowledge/education to redesign it.)

[paulie]

If you read this thread you should know electrolytic is the worst choice. Tantalum is almost as bad. Not only are these dangerous because they are polarized but they are ten times bigger and cost 10 times more than ceramic chips.

Not to beat a dead horse (aka to beat a dead horse) who's bright idea was it to use a cathode bar to indicate positive on tantalums? IMO just asking for trouble and another reason to avoid them.

[nctnico]

Not to beat a dead horse (aka to beat a dead horse) who's bright idea was it to use a cathode bar to indicate positive on tantalums? IMO just asking for trouble and another reason to avoid them.

Over a decade ago I worked at a company where the production department put tantalums the wrong way around. They didn't explode right away but instead they worked as time bombs. After the first boards went up in smoke (taking traces on a very expensive PCB with it) we had to visit all the sites these boards where installed and correct the problem. For some silly reason the PCB design guy didn't find it necessary to put a silk screen on the PCBs so it was hard to see how the capacitors need to be mounted.

[Kjelt]

Tantalum is almost as bad. Not only are these dangerous because they are polarized but they are ten times bigger and cost 10 times more than ceramic chips.

Not true, they are <three times bigger and +/- cost <two times more BUT they do have their full capacitance over their rated voltage.
And ESR is also not very bad, at 100Hz the tantalum is better than the mlcc but at 10kHz the mlcc is better than the tantalum. Just did some tests with what I had laying around:
note that you can only use the MLCC till around 9V to have some capacitance left and the tantalum is 25V so three times higher.

                                                      MLCC                  Tantalum
Voltage:                                            16V                      25V
Manuf. prov. Capacitance:                   4u7                      4u7
Measured Capacitance:                       3u4                       4u6
Measured ESR @100Hz:                      27                         11
Measured ESR @10kHz:                      0,1                     0,37
Metric Volume in mm³                         3,57                       9,7
Lowest Price (Farnell @2000 pcs)        0,04  (X7R)           0,067
Lowest Price (Farnell @20000 pcs)      0,036 (X7R)          0,046

Still I prefer the MLCC just keep in mind that you overdesign them, only use X7R or better and casing 0805 or bigger otherwise the voltage derating is shocking.

[dannyf]

So what should one do in hobbyist design?

I wouldn't worry about it - the issues here are so marginal that it has very limited impact unless you design for NASA's deep space projects,

[paulie]

For once (maybe twice ) I have to agree with dannyf.  It seems like many hobbyists like to pretend they are building critical care medical equipment or Boeing 777 black box when really just trying to get an LED to blink,

A quick update, I uploaded gerber files (2 layers, using Elecrow cam file) and ordered 10 1mm boards from elecrow. Will see how it goes. https://github.com/zapta/arm/tree/master/mini-pro/eagle

Are those for your "arm-mini-pro"? Why there?

If so I am quite interested because this looks like a perfect board for that chip. A couple friends might also like to build one and Gerbers would be helpful. Unfortunately I can't get into that page. Would it be possible to zip them up here?

[zapta]

A quick update, I uploaded gerber files (2 layers, using Elecrow cam file) and ordered 10 1mm boards from elecrow. Will see how it goes. https://github.com/zapta/arm/tree/master/mini-pro/eagle

Are those for your "arm-mini-pro"? Why there?

If so I am quite interested because this looks like a perfect board for that chip. A couple friends might also like to build one and Gerbers would be helpful. Unfortunately I can't get into that page. Would it be possible to zip them up here?

Yes, you are right, wrong thread.  Attache here are the gerbers' I sent to Elecrow  (they are are also available at https://github.com/zapta/arm/tree/master/mini-pro/eagle  ). They were generated using Elecrow Eagle CAM script.

You may want to wait until I build and test at least one board to see that the design is good. Elecrow order is with DHL shipping (~48 hours door to door) so  should be here in 10 days or so. I used the 10 PCBs deal but need only 2 so will have spares to give away, they are 1mm blue HASL boards.

BTW, the Elecrow deal is for 5cm x 5cm boards so I considering adding free 0.1" pitch prototyping area but decided to focus on small size that fits in solderless breadboards.

[paulie]

Thanks. For some reason I'm locked out of the site for "failed security".

I will definitely wait for your report then try one myself before dragging others into it. Everybody is fascinated by the prospect of drag-n-drop flash programming. This would indeed make it by far easiest and no hassle method. If the files required are true binary then that should be easy to solve by simple file renaming. I look forward to your results on that.

[gmb42]

Thanks. For some reason I'm locked out of the site for "failed security".

Click on the zip file link in the middle of the list, then on the resulting page click on the raw button and your browser will download the zip file.  No need to be logged into Github.

[zapta]

Thanks. For some reason I'm locked out of the site for "failed security".

Click on the zip file link in the middle of the list, then on the resulting page click on the raw button and your browser will download the zip file.  No need to be logged into Github.

If this does not work, you can download the entire repo  https://github.com/zapta/arm/archive/master.zip . It includes the two lpcxpresso projects (libraries and hello world) and a few resources (e.g. an appnote regarding USB ISP).

[paulie]

Thanks for the suggestions guys but maybe there is some system corruption or virus here because at github.com I get:

Secure Connection Failed
          An error occurred during a connection to github.com.
Cannot communicate securely with peer: no common encryption algorithm(s).
(Error code: ssl_error_no_cypher_overlap)
  The page you are trying to view cannot be shown because the authenticity of the received data could not be verified.
  Please contact the website owners to inform them of this problem. Alternatively, use the command found in the help menu to report this broken site.

I thought it was ok last week so I'll restore a drive image from then and see what happens. Or maybe they are playing git games again. I hate those sites. For now attaching here did the trick. Thanks again for the gerber files. If your board works they will come in handy if not I'll wait for fixes.

[zapta]

Thanks for the suggestions guys but maybe there is some system corruption or virus here because at github.com I get:

Secure Connection Failed
          An error occurred during a connection to github.com.

Interesting, are you using a recent version of a mainstream browser?

I tried to attached here a zip with the full repository but it was too big (6.5M) and got rejected.

[paulie]

I'm using firefox 19.0.2 which worked fine on github.com last week. So unless something changed there then maybe something in my system is corrupted. I've been screwing with my router DMZ and firewalls lately to control things in my house from remote locations with the new esp8266 wifi modules. That may be involved.
 
Please don't worry repository contents. I can't take advantage of those anyway because the SSD in my netbook only handles one bloatware IDE at a time and LPC or PSOC are not a high priority for that. I have a half dozen of the no-nonsense toolsets like in the $1 thread and they satisfy all current needs.

[gmb42]

I'm using firefox 19.0.2 which worked fine on github.com last week. So unless something changed there then maybe something in my system is corrupted. I've been screwing with my router DMZ and firewalls lately to control things in my house from remote locations with the new esp8266 wifi modules. That may be involved.

Nothing to do with anything other than your ancient (in internet lifetimes) browser. 19.0.2 is over 18 months old, and the the world of TLS\SSL certs and ciphers has moved on a little and left you behind.

The error message indicates that your browser and the server can't agree on the cipher to use to safely encrypt your session, probably because your browser, being so old, doesn't support any protocols or ciphers that are currently considered safe to use.

Another possibility is that your browser isn't offering to use TLS, and due to the latest TLS\SSL vulnerability, github is responsibly refusing to support anything less than a TLS connection.

Hopefully you don't use that browser for anything where safe TLS\SSL use might be important to you.

[paulie]

Thanks for bringing that up. I won't pretend to know exactly what's going on but following your advice improved things big time. Funny that everything was great a week ago. Then next few days it got so bad I had intermittent trouble with this site and finally couldn't check out at Cypress without similar warnings.

I hate updating. Seems like every time it does NOT get better. Less on the screen and funny shapes and colors with the new Firefox but maybe I'll get used to it. I remember when it was called Netscape and went years without updating. Same for IE. 4-5 years with the original XP version then things started acting funny again. Updating IE was a disaster so along came 'Zilla who chased the evil monsters away. Anyway now I can navigate better and buy stuff everywhere. Thanks again.

[zapta]

The Arm Pro Mini PCB arrived from elecrow and worked straight out of the reflow oven. Reflowing the 5mm MCU was easier than I though, it has plenty of spacing between the pads. One nice trick is that a virgin LPC11U35 starts in USB/ISP mode without having to press the ISP button which makes the initial programming very easy (just drag the binary file to the disk drive).

Elecrow sent me 10 PCBs so I have spares. If anybody wants want to play with it pm me.

Schematic and board files are here https://github.com/zapta/arm/tree/master/mini-pro/eagle

Thanks everybody for the feedback.

[paulie]

This is exciting. Very good sign that it worked straight out. Usually takes me at least 2 or 3 tries and a few hours debugging.

The ability to drag and drop strikes me as a major advantage. 90% of the problems people run into getting these things going deal with flashing and related issues. Drivers, utilities, OS, whatever.

I sent address info and if you can spare it would love to get one with just a chip on it for minimum component experiments. Based on your success and one of those Aliexpress deals I ordered 5 chips for just over 2 bucks each shipped. In fact went back to get another lot but sales is over now so maybe Mouser if these catch on at the next ham club meeting.  Maybe this particular variant not technically a member of the $1 club but worth a few extra cents if the virtual drive thing works as well as it seems. Thanks again for all the hard work.