NerdKits - electronics education for a digital generation

You are not logged in. [log in]

NEW: Learning electronics? Ask your questions on the new Electronics Questions & Answers site hosted by CircuitLab.

Everything Else » Electronic component quality

April 08, 2014
by sask55
sask55's Avatar

Recently I have had a number of IC chips fail. It has become apparent to me that these Toshiba TB6560AHQ chips can be difficult to work with. There are plenty of posts on various web sights detailing precautions or steps to be taken to try and protect these chips. Unfortunately my experience has bean that even after carefully adhering to all the power up and power down sequencing, fussing and control signal timing criteria listed by Toshiba on the data sheets and “Chip Usage Considerations” documentation I have still had problems.

These chips quite literally blow. Twice I have had a relatively dramatic chip failure for no apparent reason. A bright flash or spark with considerable sized pieces of the chip flying several feet, plenty of smoke. There is absolutely no question the chip is destroyed. The 24Volt motor power source is capable of 15A but is fussed to 3A per chip, so the available power on these chips is relatively high. The chip failures are sudden and dramatic it is almost as it there was an instant internal dead short within the chip of the motor power supply current. Prior to the failure everything was working as expected. The software on a connected PC was controlling motor speed, direction and, torque level. Then suddenly in the midst of what should have been a typical command sequence POP. Previously I had been occasionally torching the heat sink on the controller chip with my finger to determine if there was any excessive heat building up. I never found any indication of heat at all in fact they are not even warm after they blow. The event is so fast that I don’t think there is time to heat up the chip.

I am starting to wander if part of my problem might be possibly be related to the quality of the chips I am sourcing. These chips are widely available from various vendors. I have found the price can differ by a factor of 3X. The chips I have now I ordered from a inexpensive vendor from China perhaps that is a good part of the problem. On the other hand I have a lot of components ordered from suppliers in china that seam to be working just fine. This leads me to wander how it is possible to know what a good value is when ordering components. It is possible that the chips I have are fine and that my issue is related to something that I have missed in the hardware setup or software. It seams to me that I have not always got better quality by paying more.

I suppose I could assume that if I order from a north American supplier I will get a high quality genuine Toshiba component. Perhaps there may be cheep (inferior) knock offs coming out of some of the other suppliers. How would I know?

For example; I have uses Digiky.ca many times. They have a huge selection very quick delivery times and reasonable prices for many items, but shipping cost can be a factor on small orders. When I compare the Digi-key.ca price to other options for these chips the price difference is substantial. At $10.12/chip or $91.00 for 10 chips plus shipping that option is expensive. In theory I only require 3 working chips and still have a one. Considering I have destroyed 4 chip to date it would appear I may require a few spares in case I have more problems.

I am hoping to start a general discussion on how satisfied people are with various sources of components? Are cheep knock offs actually common or is a chip marked as a Toshiba chip a likely a Toshiba chip from anyone?

Just one example site of how many different options and prices are available for this chip. How is it possible to sort out what is good value?

April 08, 2014
by sask55
sask55's Avatar

The link does not apear to be working

above link

April 08, 2014
by JimFrederickson
JimFrederickson's Avatar

I have never ran into parts that seemed to have a quality issue that I have purchased from United States Companies in the United States, but then again the majority of the parts I purchase are from suppliers I have used for a long time.

Digikey has been, for over 20yrs, my favorite source.

Often I have used Mouser as well, but I got out of the habit of using Mouser when their Website worked strangely for 2+ months.

I am not purchasing most things in Quantity, and I am not getting very many expensive parts. (This is just fun for me.)

I also, as noted before, like Sparkfun and Evilmadscientist as well.

Strangely pretty much everything can be "conterfieted". (Often times I really wonder if anyone is really making money at it, but they must be because they have the capital to do it.) I have ran into that in Asia, (specifically the Philippines and Thailand).

I think it was about 2 yrs ago there was a whole chain of "counterfiet/fake Apple Stores" in China. (Most of the products sold were legitimate, but the Stores posed a Apple Stores.)

Part of China's intent, standard operations, is to Source Manufacturing for Foriegn Countries and then they produce their own "variants of the parts".

I would get a couple parts from a source that you like/trust, and then see if those work better.

My experiences with heat issues is that the chip gets hot, sometimes "smokes", then once in-awhile-blows after that. (Mosty the "smoking times" are "experiments to determine limits, but sometimes it's just ME!)

I lean toward the most likely issue you are having being a Circuit issue or Signal Issue.

Back EMF is always an issue with motors and can blow things suddenly/instantly.

For testing I would "simplify your usage". Maybe run a motor at a constant speed, (steps/sec), in a constant direction for a short time.

Then make that time longer.

Then maybe overnight.

If that all works start introducing variance and more complexity. See where/when the issue comes up.

April 08, 2014
by sask55
sask55's Avatar

Thanks Jim

The first couple of these chips that where damaged it was clear to me after the fact,what I had done to cause the failures. The last two failures where much different in that they happened doing something that had been done many times before with no problem. As you have suggested I have been making small incremental progress on this thing, off and on, for a very long time. I can't identify anything in my hardware or software that had be changed or modified in any way that may have triggered these sudden chip burnouts.

I really have no idea if the chips I have been using are in anyway responsible for this issue it is just one of the possibilities that has occurred to me. Honestly I suspect it is more likely some software timing event that I have not considered in my programming. There is a lot of independent real time events and communication going on triggering interrupts on the four Micros. information originating from the three motor controller chips and four digital calipers is arriving at the PC. Interrupt clashes are inevitable. I have been trying to gain a detailed understanding of what the consequences of any interrupt classes may be. I don't know of any set of motor controller chip input pin state or timing sequences that I could set to cause the chip to more or less blow itself up. I do not think any interrupt clash could possibly effect the controller chip other then effecting timing and possibly combinations of input pin state changes. So at this time I am trying to envision some input pin combination (on the controller chips) that could possibly cause this result. since all possible combination could be deliberately set at times to achieve certain chip movements it is hard to believe some unforeseen inadvertent sequence do to a class would result in a total chip failure like I am experiencing.

Like you I am doing this just for the fun of it, At times I have too many irons in the fire and can't find the time I would like to devote to my electronics hobby. As a result progress can be very slow.

I am ordering some more chips to continue my testing. I am have some trouble deciding if I want to order the relatively expensive digikey chips where I feel more confident in quality or the much less expensive chips. Any errors I may make may destroy the chip anyway. In any even with crop seeding just around the corner I may not actually get around to doing more testing until after the growing season this fall. Again very slow progress.

for now I am considering possible contributing factors to investigate later.

April 09, 2014
by JimFrederickson
JimFrederickson's Avatar

I am going to post a follow-up on your "Timer Interrupt Problem" Post...
(Since it will have more to do with Project/Program Structure than Component Quality.)

If I may ask, what do you Farm?
(Well raise on your Farm during the growing season...)

April 14, 2014
by sask55
sask55's Avatar

Jim

I just read your last post here now.

We grow grains, oil seed and pulse crops. specifically wheat, mostly durum wheat, Canola, dry field peas, lentils, sunflower seed, barley, flax seed and other crops from time to time over the years. This year we are trying a few acres of soya beans. We hope to be on the land seeding by the fiirst week of May or earlier if the weather cooperates.

The first ¼ section of our family farm was homesteaded by my grandfather in 1902. My youngest son and I are farming together now.

Thanks again for the help.

Darryl

Post a Reply

Please log in to post a reply.

Did you know that negative numbers are represented in two's complement notation in binary? Learn more...