hlipka

I ordered some chips from the estore on Wednesday (for about $10), and there was no shipping (at least to Germany). Should arrive today...

Under "New in this release": "License manager-free release"

And I just received the online activation for CCS It even already appears in my TI account so I don't need to enter id...

I just ordered. The store added two additional line items (for the LP and the CCS) to the shopping cart, but the total amount was still 16.98. It seems as if the promo item is just a bundle.

Switching MISO and MOSI for a SPI EEPROM, while building my PicKit3 clone. On the original schematic from the PK3 user guide they are called SDI and SDO, for both the MCU and the EEPROM. So I naturally connected SDI to the same net, and SDO to another one. But in reality SDI was the input on both sides, and SDO the output - so they should have breen swapped. It took me about 3 months to find this out

Yes, I did (and its still my preferred prototyping platform). But first I should add that its perfectly possible to use Keil, IAR or Eclipse for the software development part, so you are not bound to the editor of PSoC Creator. Linux support is still bad, though (I use a VM with XP in it) My first project, several years ago, was a combined analog-digital scope. 2 analog channels, 8 digital inputs, 1MSpS, up to 48kbyte storage (so 16 samples for everything). Digital triggers where based on bit-state with masks. Analog triggers where with configurable trigger level and slope and edge direct

The SSL demo project for the CC3200 explitely says that a connection cannot be up-graded to use SSL (nor can it be down-graded). It also mentions this as a limitation of the CC3200.

The ADS1118 has an built-in temperature sensor. Look at the documentation of the booster pack (http://www.ti.com/tool/430boost-ads1118), there you find the schematic. Apart from some passives, ESD protection and the LCD there is nothing else on the board...

The question about the distance of sensing made it into TI''s LDC1000 FAQs: http://e2e.ti.com/blogs_/b/analogwire/archive/2013/10/11/inductive-sensing-top-three-ldc1000-questions.aspx?hootPostID=82de5e49190b64ee7efb9439a28c60e3

If you see noise or interference, you can try to increase the "response time" setting (or reduce the data rate). As far as I understand it, the former should increase the oversampling (though its not really explained in the data sheet). According to the data sheet, the lowest frequency is 5kHz, so it should be possible to set this in the GUI to about 20% below the resonant frequency you measured (about 22kHz).

I got some feedback from TI about my article. One of their product experts told me that, for a more general metal detector, one either needs a larger coil or should add a ferrite backing to it. So I did some more experiments, though not with a larger coil (so the results are not as good as they should be). I need to go into the coil winding business...

Yes, I used the supplied PCB coil only. But I also looked at the WEBENCH tool for the LDC1000. When you try to calculate coils to detect 1cm copper object, you get at most a distance of about 6cm (though the tool only goes up to 7cm distance).

The data sheets states "1.7mA" plus the LC tank current. So it seems to be OK when its not always on.

I had first go on the EVM (it arrived on tuesday): http://blog.hendriklipka.de/archives/2013/09/ldc1000_test.html This sensor is really sensitive (and accurate), but it is intended for small distances. So anybody thinking about using it as a metal detector will be disappointed.

A year or so ago I bought a EKS-LM3S8962 kit, which contains a board-locked version of CCSv4. is it possible to upgrade this license to CCSv5? Or does the free license run with this board in an unrestricted manner?

I either use my trusted HP48GX (own it since school) or SpeedCrunch, when my computer is already on. The latter one doesn't use RPN, so I need to adjust to it everytime, but I can put whole formulaes into it, and I like its large history of my previous calculations.

I own a charger with a "refresh" mode. Basically it charges and discharges the battery until the measured capacity doesn't change significantly. Recently I had a battery the charger wouldn't recognize - it had a voltage of about 0.8 volts. I charged it with 15 milliamps (which is what my home-made bench power supply delivers for LED tests) for about 10 minutes. After that, the charger accepted it, and it had it's full capacity back after only 2 cycles. You might want to try this - just charge it manually with a low current until the battery voltage is high enough, and then put it into the char

According to the data sheet, the output ports have a voltage drop of 0.3 volts. So they can deliver only 1.5 volts in your case, which might be too low. Can you check with a multimeter that the ports are active at all? Have you checked that the output of the LM317 really is 1.8 volts? And what do you mean with "50 hz wave in each LED"?

You're regulating 12 volts down to 3 volts. Thats 9 volts difference. Let's assume you draw 20mA (I don't know exactly how much power a Launchpad draws, or all the logic you need to power from 3 volts). This means you need to convert 180mW to heat (9V*0.02A). The LM2937 in TO220 case has, according to the datasheet, a thermal resistance of 53K/W (page 8), meaning its temperature will rise about 10 degrees (celsius) above ambient. If you draw 40mA, it will rise 20 degrees. 100mA would mean nearly 50 degrees above ambient, enougt to burn your fingers. (The TO-263 and SOT223 cases have an even hi

It's a difficult question. I really liked the quick startup of the cloud solution - it took about 30 seconds to get every thing up and running. And I still like it when just doing a quick experiment or test. But when you build a more complex software, maybe build of multiple modules, having more control is really fine. Fortunately they now support downloading all your stuff to your machine, so I can build with CodeSourcery. best of both worlds (and they now also support version control in the cloud, which makes many things easier).

* two CY8CKIT-050 (PSoC5 DevKit) * a CY8CKIT-001 (PSOC5/3/1 DevKit) * STM32F4 discovery kit * STM32LVDiscovery kit * EK-LM3S8962 (CAN+Ethernet) * a ezChronos watch * some LaunchPads * ChipKit32 Max * EFM32 gecko STK * EZ430-F2013 * F28069 Piccolo controlSTICK * a mbed

Mill-Max also ships samples (in small quantities).

I think Maxim limits email addresses to .com and .org (and maybe .edu). Last time I used my company email.

Just installed it - there is a drop-down box for the chip series. it contains the G series, then you can select anything up to the 2553. (did not test it though)

Got mine too today. The packaging is hilarious - a Fedex medium box for stuff which would easily fit inside a small envelope I too was disappointed by the BGA package of the Stellaris MCU. This was the part I would have liked to use first, and was hoping for LQFP packaging there... Whats also strange: the National page for the LM5101 states it as 'not recommended for new designs'.