Data Recovery: Kingmax Pro 32Gb microSDHC – A Sign of TLC Weakness?

If you think I’m not having much luck with microSD cards lately, you’d be right. The first issue for the year was the Sandisk 128Gb microSDXC card having some unreadable sectors causing the loss of some files. Now, I’ve had this Kingmax Pro 32Gb microSDHC I reviewed in the past cause issues as well.

The Issue

This particular Kingmax Pro 32Gb card was used in a Raspberry Pi, which was kept in the loft area of the house (warm temperatures) along with another Raspberry Pi with a Sandisk 4Gb Class 4 card. Both units were hooked up to a single power supply and were running various radio-based experiments.

Around a year ago, for power saving and because I didn’t require the units to keep operating, power was disconnected and both units remained untouched in the loft.

After salvaging the units, I decided to read out the home folders of both units, archive the data, and reuse the cards for another application.

The older Sandisk 4Gb Class 4 card completed its read-out successfully with no errors. The Kingmax card initially refused to be detected, and required three insertions just to get it to detect. The partition table was read but the rest of the unit could not be accessed.

I decided to let HD Tune Pro try each sector sequentially – it took over 7 hours to have the program “overflow” with errors and abort itself with nothing but the first few megabytes readable.

Not feeling defeated, I removed the card and then plugged it back in and tried again. Miraculously, everything now was able to read out – but only very slowly, taking 1 hour 41 minutes and 42 seconds for 32Gb – an average rate of 5.24MB/s, well below the Class 10 rating of the card. Feeling a little more adventurous, I cycled the card again …

This time it took just 6 minutes 28 seconds to read and has recovered its full speed. No data was lost, and the read-out data appeared to be correct.

Discussion

It seems to me that this behaviour seems to suggest a similar phenomenon to the Samsung 840 Evo SSD problem with slow data access for “old data” stored on the planar TLC NAND flash. It seems that low-cost memory cards are made with planar TLC, and aside from having weaker write cycle endurance, it seems to have lower retention endurance as well.

The issue is with TLC recording, more voltage levels are used, and the read-out is more sensitive to loss of cell voltage. The leakiness of flash seems to have increased with smaller lithography – which is increasingly used to reduce costs of flash memory by reducing silicon usage/increasing yield per wafer. Combined together, retention times of about 10-years formerly claimed are now realistically probably about 2-3 years based on my own experience.

This card was only unpowered for a year, but the difficulty in reading out the data suggests it was close to losing its data. The fact the data “returned” suggests the card’s internal controller was busy retrying and recovering the data, and internally rewriting the worst eroded data first. By the time I had the first complete successful read, it seems likely that the worst eroded data had been fixed, but slowdowns remained due to some weak data that could be recovered “in time” before the card was to issue an error. Post-recovery, speeds increased to normal as the data was likely all rewritten – this consumes a write cycle of its own and with increasing wear on the flash, increases the leakage rate.

An exacerbating factor may have been the slightly warmer environment of the loft area – however, it’s good to note that the Sandisk 4Gb card also used in the same way and stored in the same way did not lose enough of its charge to cause readability issues. This may be controller related in some way, or it may be because it was an older card based on MLC or larger pitch TLC flash.

Regardless, it seems inadvisable to use “modern” consumer grade flash memory for anything more than as a “temporary” storage vehicle for shuttling data around. Long term retention is not its strong point and the price-sensitive TLC units appear to make situations worse.

Conclusion

At least this Kingmax Pro card ended up with a happy ending in that no data was lost. However, it seems to confirm a glaring weakness in cheap, consumer-grade TLC memory – especially those destined for low cost microSD cards or USB sticks. Not only do they have a limited write cycle endurance, they also seem to have data retention endurance problems. Leaving them unpowered for a year seemed to be enough to cause data recall difficulties – chances are, if left unpowered for about two years, maybe the data would all have been lost and the card itself “damaged” due to the loss of internal configuration data.

It points to the whole issue that flash memory is not good for long term data retention. While formerly, retention would commonly be rated for 10-years, whether this is possible under TLC and smaller lithography processes is debatable. The industry seems to point to one year retention at “end of life” as being the norm.

It’s a good time to stop and think about all those special event photos and videos being delivered on USB flash disks and how long they will remain readable for. It’s probably best to back them up to a diverse range of media, and continually “transfer” them to newer media as time permits to stop the “erosion” of time.

Posted in Computing | Tagged , , | 1 Comment

Teardown: RadioShack Amplified Telephone Earpiece (43-229)

After my experimentation in building a T-coil receiver on my own, it seemed that the unit did have some issues which I eventually corrected. While it seems the schematic may have been a victim of a transcription error, I thought it would be nice to share the inspiration behind the whole project and where it’s gone since.

Inspiration

When I was younger, back in 2004, I worked at the Blacktown Tandy store for work experience. It was an interesting time, because already then, the store was in a “shake-down” phase as there was a local Dick Smith nearby and they were headed towards an eventual closure. It eventually happened, although since the Tandy performed well, it starved off its death for a few years. In that time, they cleared out the “InterTAN” RadioShack stock which they imported, and replaced them (mostly) with local Dick Smith products which were cheaper.

One of the good things about this was that there was a number of products which they held stock of, and nobody bought. Things like 100m rolls of 210mm and 216mm wide thermal fax paper, which they cleared out at just 50c a roll. I bagged a number of them (probably 10), and then respooled them into separate ~50m rolls so that the machine could take it, before forming a “continuous feed” system with the roll outside of the machine instead. Of course, I never received many faxes, but instead, I used a 9v battery to form the loop current between the machine and modem, and used the Sharp fax “pick up” number of 5** to trigger a transmission between a computer and the fax machine without any need for ringing current. This worked to turn the fax machine into a slow printer, which I used to print a number of reference documents for my own use. It was good value.

Another thing they got rid of was this amplified earpiece for clipping onto a phone using an adjustable silicone rubber band. Used by the hard of hearing, it would boost the telephone receiver volume, but it had no microphone on the rear. At the time, I thought this was somewhat magical – how could it do such a thing without a microphone. I didn’t realize that if it had a microphone, a feedback loop would be inevitable.

Of course, when it was supplied, it wasn’t yellowed and didn’t have wires poking out of it. But that being said, they got rid of them for about $2 a piece, so I couldn’t resist buying a few, although I managed to lose every other one.

One of the reasons this unit was a poor seller was the use of the “odd” N-size cells which have never been very common around here. They’re also fairly expensive to obtain even where available, so I couldn’t put this to use immediately. Its destiny was modification.

The first thing I did was solder to the battery terminals and hook a regular AA cell to it. I slapped the unit onto a phone and it did work – it managed to receive the audio just fine and amplify it. I started putting it on a number of devices and found it could receive audio or some odd noises as well. By then, I realized it was probably a magnetic coil pick-up with a speaker behind, and I’ve always wanted a magnetic coil pick-up.

Part of the reason for wanting one stems from wanting to be able to record the telephone. I was born too late to be a phone phreak, but I had some modems I really wanted to record. Much better methods were devised, including using a voice modem as a monitor, and eventually just using VoIP packet interception, but at the time, this was what I had. As a result, I cut-out the speaker as well, and soldered a pair of wires to plug into the line-in of a sound card.

Seeing as the unit doesn’t get any real use, but was the inspiration behind the whole induction loop receiver idea, I felt it would be befitting that it would be taken apart for a quick peek at the circuit.

Part of the secret as to how this works is on the left. The green and white wires come from the magnetic coil pickup which sits underneath the shaped plate. The plate is made of a particular sort of metal that resists magnetic fields, similar to those used to shield hard drives. This “breaks” the potential feedback loop between the speaker (removed) and the pick-up coil. The PCB itself is a paper-type single-sided through-hole board with a combination power and volume rotary potentiometer. Judging from the trace shapes, it could have been designed by hand.

There is white silkscreening on the top, and the code suggests the PCB was made in Week 34 of 2000. The circuit is made of all discrete components – no op amps to be found here, but instead four transistors, and an assortment of resistors and capacitors. The black and white wires represent the audio output.

Here’s a look at the PCB from many different angles so you can see the components that were used and their values. I don’t have the time to trace out a schematic, however, I’m not even sure that identical components are available. I suppose it’s probably possible to substitute more modern transistors in their place.

That being said, I remember that the audio wasn’t so clean, instead having some hiss on the output maybe due to the amplification intended to drive a speaker. I’ve never tried operating it outside the house, although I suspect its performance on a train may not be so good, as the shielding plate may shape its reception to be more close-field for a nearby speaker rather than for a hearing loop transmitter coil which could be more remote. However, I have not tried it, so I really don’t know.

It’s interesting to see how often you have these kinds of “weird” ideas pop up early on, but at the time, I was only just starting out in electronics and only had basic soldering skills. It wasn’t until about 10-years later that I could build something of my own that did something similar, if not better.

Project Evolution

As I wasn’t entirely satisfied with the quality of the recordings I obtained from the receiver I had built, I wanted to seek out better designs. Unfortunately, there weren’t many, however the one from Silicon Chip Magazine seemed to be the only one out there. These designs often made their way into kits, such as this one from Jaycar or this one from Altronics. Unfortunately, the price was too high for my liking from Altronics, but I still wanted to know exactly what they did. I tried to succumb to ordering from Jaycar, but surprise-surprise, they had absolutely no stock.

As a result, I went on a wild goose chase looking for the September, 2010 issue of the magazine. Sadly, a number of libraries I had checked do not hold stock of the magazine, and ultimately, I had to register with the State Library of NSW to obtain access.

As the publication is copyrighted, I can’t share the schematic with you. However, you can probably go seeking for the issue if you want to know just how they did it. Some more generalized hints as to the main part of the circuit include the use of a number of stages involving a TL072 dual JFET Op Amp amplifying-and-filtering with each stage followed by an LM386 audio amplifier. The pick-up was a xenon trigger transformer.

After seeing their design, I felt it was rather logical, so I ordered some components and set about building an abridged version of it. I felt no need for their LED power indicator which was likely to suck as much power as the circuit itself, and I felt no desire to use a 9V battery. I had no need to drive headphones directly, as I had a fairly sensitive audio recorder I could employ. As a result, I removed the LM386 entirely, and the zener/LED combination. I had no xenon trigger transformer, so I opted for a similar inductance instead. I decided to build it in the Hammond enclosure I received as a sample, to make it a little less fragile.

I opted to make things nice and neat … from the outside.

But from the inside, it was a bit of a patchwork. Note that this wasn’t the final configuration – there was still more wire cutting and jumpering around as I noted the level control didn’t do as much as I had intended and instead used it on the output instead of the second stage. This instead had the side effect of introducing pot-crackle when the level was adjusted, so … I think it’s probably better to omit it altogether and have the level on maximum at all times, but it’s too late for that now. Most of the inside volume is batteries – two AA’s to be precise.

I can’t speak for the performance of the original circuit as designed, however, this one does perform about equally well SNR wise as to the unit I built before, so it was a bit disappointing to see that all that effort didn’t result in any tangible audio quality gains. That being said, it’s good to know that too, since it shows that the EMI is probably limiting the SNR and thus, improvements in noise level or intelligibility are likely only to come about with DSP noise reduction (and its associated issues in terms of audio quality).

Conclusion

It’s been a few years since I did the initial project, but the impetus to revisit it was due to discovering the unit that started the whole idea in the first place. After a quick revisit by building one based on the Silicon Chip Magazine design, it was discovered that the audio quality limitations are likely to be inherent to the transmission system, rather than of my own design. It also seems I’ve made some probable transcription error or drawn an earlier revision of my design prior to a number of fixes, as pointed out by a reader. Unfortunately, since the unit is under a mountain of stuff now (including a pile of hot melt glue), I probably won’t get around to revising it – however, if you’re in Sydney, you can probably hop down to the State Library and grab the Silicon Chip design yourself.

Posted in Audio, Electronics | Tagged , , , | 1 Comment

Teardown: FiiO E5 Headphone Amplifier

The FiiO E5 was a low-cost but decent performing compact headphone amplifier a few years back. The unit bore a vague resemblance to an iPod Shuffle clip-version, and was relatively compact with integrated rechargeable lithium-polymer battery. Since someone didn’t want theirs, I managed to snag it second hand for cheap, but I never really had that much of a use for it. After a while, the clip broke, the battery wasn’t really lasting and I decided I might as well tear it apart and throw it out. So here’s a few pictures which I took much earlier this year, which I never got the chance to post until now.

Teardown

The exterior is a nice aluminium curved shell, and the interior unit slides out after some fiddling with the clips and persuasion to break the glue used inside. The PCB sits inside a plastic frame, with its rear littered in test-points which is covered by a self-adhesive layer of plastic. There is a terminal mounted at the bottom, which provides the grounding point to help reduce any RF interference which might be experienced by the unit.

The internal metal has been cleared of paint where the grounding contact would sit, which is necessary to make it effective.

All the magic is on the other side of the frame, where space seems to be very well utilized. The Li-Poly battery can be seen up-front, with the headphone jacks (in/out) being the next highest profile components. There is a mini-B port for charging.

The battery is marked AK402030PL and has a printed capacity of 190mAh.

The main PCB has all its components mounted on the one side, and all of them are surface mount. Quite a few ceramic capacitors, a tantalum, a number of resistors, two LEDs for charging/operation status, three push switches for volume up/down and power, and a slide switch for bass boost. It seems that the unmarked 8-pin IC is the controller that runs the whole unit. The Burr Brown/Ti OPA2338UA dual op-amp is likely to be responsible for buffering the input and doing the bass boost function. Finally, the TPA6130A2 138mW DirectPath Stereo Headphone Amplifier with I2C Volume Control does the actual driving of the output. It seems there’s a 5-pin charge controller/voltage regulator chip for the li-poly cell marked 54b9, although the battery does have its own protection board.

Conclusion

The E5 did indeed contain the promised Texas Instruments solutions in the signal chain, which explains why it was considered to have good performance, but it was made by the Chinese and at a very low price. This was all good for the consumer, as decent performance could be had for cheap, and it seems that the more savvy consumers know to look towards these brands as they develop a positive reputation.

Posted in Audio, Electronics | Tagged , , | Leave a comment