Review: Panasonic/Matsushita UJ240 6x Slim SATA Blu-ray Writer

While most people have given up on optical media (and I don’t blame them), as an early adopter of Blu-ray, I continue to stay the course somewhat. Part of that involves getting compatible drives for my gear. I recently purchased an refurbished Asus laptop, which I was very happy about, but the internal drive supplied was a PLDS DVD+/-RW drive.

Luckily, such items are generally upgradeable, and a Panasonic/Matsushita UJ240 6x Slim 12.7mm SATA Blu-ray burner was available for a very reasonable price of AU$72 from eBay. It took a few weeks before it got here in a plain cardboard box, but I was so excited that I had already swapped the face plates and screwed the attachment tongue to the drive!

DSC_7646

This particular drive was manufactured March 2013, so it’s a while back. It seems like inventory doesn’t move very quickly, and that might just be down to a slow PC market and the lack of users demanding such high specification optical drives. This drive was a UJ240 of hardware revision 1.01 and firmware revision 1.01 (fairly early, it seems some earlier drives have later firmwares as well, which is interesting).

The drive has been warranty sealed by a Chinese PC distributor it seems, but the top label also mentions Fujitsu Technology Solutions, so this may have been a FRU for a Fujitsu laptop sold as a “bare” unit.

DSC_7650

The original flat face panel (for very old fashioned laptops) is emblazoned with all the applicable logos. The drive itself features a (barely adequate) 2Mb buffer, and is detected by applications as MATSHITABD-MLT U with no firmware revision.

From what I can ascertain, this drive is specified for maximum write speeds of:

  • BD-R: 6x CAV for Single Layer, 4x P-CAV for Dual Layer
  • BD-RE: 2x CLV for both Single and Dual Layer (some sources claim 4x, but no 4x BD-RE exists to my knowledge)
  • DVD+/-R: 8x Z-CLV
  • DVD+/-RW: 4x Z-CLV
  • DVD-R DL: 2x CLV (some sources claim 4x)
  • DVD+R DL: 2.4x CLV (some sources claim 4x)
  • DVD-RAM: 3-5x Z-CLV
  • CD-R: 8x CLV (some sources claim 24x)
  • CD-RW: 8x CLV (some sources claim 16x)

It seems some sources may have got read and write speeds mixed up, or there are firmware differences between drives. As a result, it’s likely not to be compatible with the fastest DVD-RAM, DVD+/-RW or CD-RW discs. The available specs don’t make it clear, but it is capable of burning and reading LTH discs. Compared to a dedicated DVD+/-RW drive, the CD operations seem rather slow, as most are capable of 24x CD-R recording.

DSC_7647

The drive feels slightly weightier than a regular drive, and upon looking at the laser carriage, it’s obvious there are two lenses – one is used for the DVD and CD operations, with the other used exclusively for Blu-ray discs.

But how well does it write? It’s an interesting question to ask, as historically, my experience with slim type drives is for very poor write quality especially at higher speeds because the Z-CLV strategy creates step-wise changes in burn quality with issues very common around speed transitions.

Blu-ray Write Quality Assessment

I only have a limited selection of media around me right now, unfortunately. I haven’t bought new blanks in ages, and I’ve probably spent more in testing than I have in serious writes over the past half-year. But I suppose you gotta break a few eggs to make an omelette … just that this guy breaks more than the average number :).

PRODIS-CR0 4x BD-R

These are cheap discs, which work surprisingly well with my gear. The drive detected the Prodisc blanks and offered to write them at 4x. The first disc was written successfully, and the drive itself read it back flawlessly at 6x.

Transfer_rate_mats+mats+pro1

Asking the Lite-On iHBS312 what it thought about it, it felt the disc was burnt horribly. The key values should be an LDC Average of 13 or less, and a BIS Maximum of 9 or less for a good burn, although I did establish this was overly harsh, and read errors were only likely for BIS values >40-50 or so.

Disc_quality_mats+pro

This is worse than all of my “ageing” Blu-rays written by old burners. Needless to say, cross-compatibility checking with my Lite-On iHBS312 showed difficulty in reading.

Transfer_rate_lo+mats+pro

A cross-check was also made with my most recent acquisition, a Pioneer BDR-209DBK (more on this in another post), which also had difficulties.

Transfer_rate_pio+mats+prod1

Cross-checking with my old and reliable LG GGW-H20L revealed that the drive mis-identifies the disc as a video disc and applies a rip-lock to it, but the drive had only minor hiccups when reading back at a leisurely 2x.

Transfer_rate_lg+mats+prod1

Thinking this was a one-off media issue, I risked a second PRODIS-CR0 at 4x, and this time, it seems there was a few tiny hiccups on read-back on the burner itself.

Transfer_rate_mats+mats+pro2

Alas, the Lite-On thought no better – in fact, this disc was worse. Damn.

Disc_quality_mats+pro2

It was so bad that there were read issues with the disc on the Lite-On.

Transfer_rate_lo+mats+pro2

There were no read errors with the Pioneer drive, although many difficulties. It is clear, Z-CLV strategy is thy enemy as the 4x zone is poorly written.

Transfer_rate_pio+mats+prod2

The LG had issues with it too … even at 2x.

Transfer_rate_lg+mats+prod2

It’s clear that this drive is hopeless at writing PRODIS-CR0 at 4x with acceptable quality, so I spent a third disc at 2x to see just how it would perform.

Transfer_rate_mats+mats+pro3

The burning drive had no issues with the disc, as has been the case throughout so far.

Disc_quality_mats+pro3

The Lite-on agrees that the burn quality is now excellent, which is what I would have expected.

Transfer_rate_lo+mats+pro3 Transfer_rate_pio+mats+prod3 Transfer_rate_lg+mats+pro3

All three cross-checkers had no problem with it at all either. It looks like the burner is only good for 2x for PRODIS-CR0.

CMCMAG-BA5 6x BD-R

This particular media doesn’t seem to work well with Lite-On drives, as I established earlier. Lets see what the Panasonic makes of it.

Transfer_rate_mats+mats+cmc

The drive happily wrote the disc at the rated speed of 6x, using a CAV strategy. The readback with the burner had a tiny blip, but nothing too major. Lets see what the Lite-On thinks …

Disc_quality_mats+cmc

Surprisingly, the Panasonic seems to like this media and burns it better at 6x than the Lite-On could ever manage. So, if you have a Panasonic slim burner, buy these!

Transfer_rate_lo+mats+cmc Transfer_rate_pio+mats+cmc Transfer_rate_lg+mats+cmc

All three cross-check judges agreed that the burn was good!

UMEBDR-016 6x BD-R

A bit of a surprise contender, since I don’t have many of these, but generally these fared well with other burners. Lets see if the Panasonic likes these …

Transfer_rate_mats+mats+ume

The disc was detected and only 4x was offered. The readback from the burner is perfectly fine, lets see what the Lite-On thinks.

Disc_quality_mats+ume

It’s another dangerously marginal burn, which exceeds our normal criteria but isn’t initially unreadable.

Transfer_rate_lo+mats+ume Transfer_rate_pio+mats+ume Transfer_rate_lg+mats+ume

All three cross-check drives only reported minor undulations in the read speeds with no serious slow-down, but it’s a sign of a poor burn. Maybe burning at 2x will improve it, although I don’t have enough of them to “spend” on this burner right now.

Taiyo-Yuden That’s TYG-BDY05 LTH 6x BD-R

While LTH media offers no great compelling benefits to the end user, these discs were purchased as a curiosity, and burnt well to extremely well. The Panasonic drive said nothing about LTH, so lets give it a try.

Transfer_rate_mats+mats+tyg

Whatdya know? It detected and offered to burn at 6x, and it came out with only a tiny glitch when read out by itself.

Disc_quality_mats+tyg

It seems, despite using a CAV strategy, the burn quality isn’t very impressive with the errors starting to get a little high on the outer edge of the disc. Maybe 2x is a better choice but I am not going to spend another disc to find out.

Transfer_rate_lo+mats+tyg Transfer_rate_pio+mats+tyg Transfer_rate_lg+mats+tyg

The cross-check readers show minor undulations, but nothing severe.

VERBAT-IMv 4x LTH BD-R

Why not throw another LTH at the drive and see how it does? This is another quality disc, so it should do well …

Transfer_rate_mats+mats+verb

It detected as a 4x disc and interestingly chose to burn using a P-CAV strategy rather than a Z-CLV strategy, which is faster and normally preferable! The readback on the burner itself looks happy.

Disc_quality_mats+verb

The result from the Lite-On seems to suggest the burn could be better. It’s a pretty average result, but by far, the best of the 4x burns so far. It does have blocks and bunches of error spikes, which suggests the power control of the laser might not be working as well as it can be. Peak BIS errors exceed 9, so it cannot be considered a high quality burn.

Transfer_rate_lo+mats+verb Transfer_rate_pio+mats+verb Transfer_rate_lg+mats+verb

None of the cross-check readers encountered any problems, however, as can be expected.

Other Caveats

Well, for one, I haven’t examined the quality of writing other types of media – say BD-RE, DVDs or CDs, so that remains an unknown. But I suppose with the data above, you’ve probably already made up your mind that this is a pretty average-to-poor BD-R burning solution.

It was only after performing all of the tests that I realized there is quite a big number of people who despise Matsushita drives because of their unusually strict compliance with RPC region controls, which makes most of their drives permanently region change limited and also incapable of software-based brute-forcing of CSS. An explanation for why this is the case is given here. Firmwares are also rare and far between. If you like to travel, this is probably not the drive for you. Unfortunately, this drive looks as to have been tested before shipment and that meant someone set the region code to 2, whereas I live in region 4, so I lost one region change out of the box.

Conclusion

At a glance, the Panasonic/Matsushita branding inspires confidence. Then, as soon as you begin using it, it gets shattered. The discs that come out of the drive seem to vary significantly, and the Z-CLV wreaks havoc. I haven’t seen such marginal burns before, even from my oldest LG GGW-H20L. Burning at 2x seems to be the safest option with this drive, and even then, I have some reservations.

I suppose, if you have one, it’s a good reader. But it’s also a terrible writer. You need to stick to either 2x CLV where it seems to do acceptably, or use 6x CAV (if your media supports it, and even then, not without risk). 4x Z-CLV is just trouble on the media tested, but maybe with more expensive media, it could manage some acceptable burns at 4x Z-CLV.

Or just buy something else. Maybe an LG for around the same price. Or a Samsung for a little more. Or better, a Pioneer if you can find one.

Posted in Computing | Tagged , , , , | Leave a comment

Tested: Two Secure Digital (SD/SDHC) to CompactFlash (CF) Adapters

It seems that the modern trend towards smaller and faster has led to the CompactFlash card being relegated for use in industrial and professional high-end photography only, where its larger size is not a problem. Most consumer electronics have since transitioned to SD or microSD cards, with the in-between miniSD pretty much disappearing from the market.

CompactFlash was a very important milestone in the flash memory industry. Being “intelligent” and having an internal controller to manage the flash (unlike SmartMedia), it was able to overcome defects, and extend capacity and speeds throughout the generations (mostly) without losing backwards compatibility. It was also led by consortium, and was not limited to just one player, which helped it become popular standard with many vendors of cards and compatible devices.

The cards could operate in one of three modes, and the True-IDE mode was popular as it allowed passive adapters to connect the cards to IDE interfaces as use as a rudimentary “slow” SSD before high-performance SATA SSDs became popular. Many high performance cards used MWDMA and UDMA modes which are common with speedy IDE hard drives to achieve high speed. Passive adapters were also available for use with PCMCIA interfaces.

Unfortunately, CF cards are starting to become less easy to find, and by extension, high capacity and high performance CF cards often command twice the price of a similar performing SDHC/SDHC card. Many users are likely to have many SDHC/SDXC slots to use cards in, and laying down a significant amount of cash buying CF cards with the knowledge they won’t be able to use it with their next camera doesn’t seem to make sense.

The Solution? Adapt.

It turns out that this was the exact conundrum I was facing in late 2011/early 2012 when I was considering upgrading from my Canon EOS 400D which used CF cards to something more modern. The EOS range had already decided to go over to SDHC, so I knew the future for CF, at least on consumer oriented devices, was limited. I really didn’t want to spend more money buying CF cards knowing I won’t have much use for them later on.

DSC_7526

Instead, I spent about $15 each on two different adapters that were on the market in late 2011. It seems that these products, possibly revised, are still on the market today. The one on the left is the “extreme” adapter, the one on the right is an unbranded “blue” adapter.

Both adapters feature a full size SD card input, and I specifically purchased them as they supported the SDHC format, so as to be sure they were capable of using cards up to 32Gb. The slot on the extreme adapter is the “push to eject” type, whereas the blue one on the right is a friction-fit type, and pull to remove.

DSC_7527

The rear of the extreme adapter has a very generic label, with uneven line lengths, and claims to be made in Taiwan. The blue adapter has no label on the rear.

DSC_7528

DSC_7529Both adapters are Type II, which is of 5mm thickness. This thickness was originally designed to accommodate microdrives, and not all devices will have slots that can accommodate it. The Type I thickness is what most regular CF memory cards are made in, and is 3.3mm thick.

There are newer designs which are Type I (thin) microSD based adapters. There are also some with multiple slots and a type of striping between the two cards for performance, however, it seems the cards cannot be read outside the adapter.

Funnily enough, the Transcend RDF8 card reader that I use most has a slot cut-out that can only accommodate Type I cards. Removing the PCB from the shell allows for Type II cards to plug in, and thus it was tested without the external casing.

Performance Testing

What we really want to know is just how well these adapters work.

Extreme SD/SDHC to CF Adapter

Testing of the adapter focused first on read performance and the types of cards accepted. Interestingly, using HDTune Pro, we determined that the adapter works for SD (not pictured), SDHC and SDXC cards. It was tested up to 128Gb and performed correctly!

hdt-extreme-sandiskext hdt-extreme-toshibaexceriahd hdt-extreme-toshexceria hdt-extreme-kingston128

In testing the performance, I have chosen the best performing cards from the database to test with. There appears to be a compatibility issue with the Sandisk Extreme microSD card, which results in a low rate of just 5.2Mb/s regardless of number of re-insertions. The rest of the cards seem to be capped off at 12.2Mb/s, which is a lot slower than the cards are capable of.

This is only a speed of about 83x, which when compared with CF cards that can perform at 133x, 200x, 266x, 300x, 500x, 600x and above, it’s not extreme by any means.

A CrystalDiskMark test was run with the Toshiba Exceria Type 2 64Gb SDXC card, which resulted in a rate of about 12.70Mb/s both ways. It’s clear the adapter is hampering performance.

cdm-extreme-toshexceria

A check using H2TestW showed no corruption, which is a good sign. The speed is a consistent 12Mb/s both ways.

h2tw-extreme-toshexceria

Blue SD/SDHC to CF Adapter

Testing the blue adapter seemed to run into some trouble. It was quite picky with cards, and had difficulty with the Toshiba Exceria HD 32Gb microSDHC and my Samsung 32Gb Class 10 SDHC card, claiming it was unformatted and was not able to complete the format. It did, however, work fine with the Sandisk Extreme 32Gb microSDHC. No SDXC cards worked in this unit.

hdt-blue-sandiskext

The transfer rate was an even lower 9.1Mb/s which translates to 62x which is below even the basic 66x cards sold ages ago.

cdm-blue-sandiskext

The performance is not consistent either, with writes showing half the speed. This is highly undesirable, especially when used for photography with mid to high end DSLRs.

h2tw-blue-sandiskext

No corruption was experienced, however, similar speeds were shown.

Other Caveats

Retro-computing users might have particular interest in these cards as CF cards start to get rarer as the true-IDE mode should allow these cards to be used in place of less reliable rotating-platter style hard drives. Testing the two adapters provided rather interesting results.

The extreme adapter identified in the BIOS incorrectly, with cards larger than 8.4Gb always identifying as 8.4Gb (16383/255/63) CHS mode. However, once booted into Windows, the full capacity is accessible and usable in True-IDE mode.

HDTune_Benchmark_SD to CF Adapter HDTune_Info_SD to CF Adapter

The unit performs slightly faster, but is limited by its MWDMA2 mode (16.6Mb/s maximum). It identifies as SD to CF Adapter with firmware version 45.0.235.

With the blue adapter, while the adapter and the size are correctly identified in the BIOS, all cards tested with it claimed to need formatting, which could not actually succeed.

unabletoformat

However, reading from it, and performing read benchmarks were possible.

HDTune_Benchmark_Memory Card Adapter + Sandisk Extreme HDTune_Info_Memory Card Adapter

The unit identifies itself as Memory Card Adapter with firmware 67281306. It seems to implement a very high UDMA4 mode, however, the BIOS has limited it to UDMA Mode 2 (33Mb/s) because of cable/adapter limitations.

deviceman_blue_udma2

So despite having the most desirable mode for a hard disk replacement, the adapter itself is still very slow and will not function correctly at least in LBA mode. It seems the true IDE mode has not been correctly or completely implemented. I also double-checked the write protection switch on the SD card, and verified it worked properly with the reader before and after a failed test internally in a CF to IDE adapter PCB.

Photography users might get rather concerned as well, as using these with my 400D occasionally resulted in Card Error, which could be resolved sometimes by removing the card and re-inserting it (and losing a few photos in the process) or occasionally losing the whole filesystem altogether and requiring file signature analysis for recovery. It seems it’s not a good candidate for speed or reliability.

Another thing that might be worth worrying about is power consumption. These adapters get fairly warm, so I suspect they may be consuming more power than most CF cards, and will shorten your battery life.

Conclusion

Using adapters may seem an obvious solution to a problem, but in this case, it comes with many caveats. The units I had displayed puzzling behaviour, with compatibility problems and inconsistent performance. Performance in general, was very poor, even when working correctly.

I find it hard to recommend because it seems hard to get a reliable working set-up without bumping into random errors here and there, and due to the numerous caveats which depend on your application. It seems it is no substitute for buying a quality, even if low-rated CF card.

The market has, likely, moved on a little since I had purchased these adapters, but you might want to ask yourself whether it’s worth the risk … I don’t think it is, so I probably won’t be trying more of these.

Posted in Computing, Photography | Tagged , , , | Leave a comment

Review, Teardown: Lanu Power LP-401B 4400mAh Power Bank

Here comes another power bank thanks to an anonymous donor. This particular power bank is a Lanu Power LP-401B 4400mAh power bank, packaged in the same package with the words “Quality Assurance” as the LP-406B reviewed earlier. As a result, it confirms the suspicion that the outer package is “generic”.

DSC_7399DSC_7400

The power bank itself was black in colour and roughly a rectangular brick. The top of the power bank has a push button at the centre of the ring which activates the power level LEDs and turns on the power bank. Long-pressing on it activates the 5mm LED torch, and toggles between on, SOS and off.

DSC_7401

DSC_7402The product is shipped inclusive of a user manual leaflet and a basic charge-only USB micro-B cable.

The rear of the power bank lists its specifications, which lists its capacity as 4400mAh and the input and output both of 1A capacity. This is suitable for most smartphones, but is sub-optimal for heavy loads such as tablets.

 

 

 

DSC_7403

The top of the power bank has its single USB A output, a single USB micro-B input and a 5mm white LED for a relatively weak emergency torch.

Teardown

The power bank is constructed by clipping two halves together. Prying at the seam with a flat-head screwdriver allowed for the halves to come apart.

DSC_7515

Internally, the two 18650 cells were shrink wrapped together into a battery pack, which is nice and neat and what you would expect from a quality pack. However, there are no markings on the exterior of the pack, and my nose detected something suspicious about this – was it hiding something?

The construction of this power-bank is similar to others in using two interconnected PCBs stacked one on top of the other. Visible is the top of the top PCB which has the surface mount blue LEDs arranged in a circular pattern to match the top. The top PCB is marked LP-601B V1.2-A and LP-401B V1.2-A, which suggests that the PCB is common between two models of power bank. LP likely stands for Lanu Power, which is a branding belonging to Shenzhen BlueTimes Technology.

DSC_7516

The underside of the top PCB suggests the PCB is made in Week 33 of 2012. Just like most other power banks, there is a microcontroller with the markings ground off. The white LED is mounted on this side, as is the connections to the battery. The soldering is positively average, with it being just quickly tacked on, with a splash of solder to the right. It’s a concerning design where the current from the battery has to travel through the connectors and a fair amount of PCB to get to the inductor, as the added resistance reduces efficiency.

DSC_7517

The bottom PCB has an open bi-filar wound inductor, which is not the most efficient sort. There is an SS34 3A Schottky rectifier diode, and a small tantalum capacitor. No bulk electrolytic capacitor is used to smooth the power output. There is also an IC marked 5056, which is likely a USB charger IC. It seems this bottom PCB is specific to the LP-401B model.

DSC_7519

The underside of the bottom PCB has the very familiar 8205A MOSFET, a few resistors to set the USB charger detection for Apple products, and a few test points to monitor input and output voltages.

DSC_7518

The battery pack was double-sided foam taped to the rear of the casing and took some effort to remove. Both ends of the pack was insulated with dark green cardboard.

DSC_7521 DSC_7520

The heatshrink had to be cut to get into the pack and evaluate the cells.

DSC_7522

Looks like genuine Samsung ICR18650 2600mAh cells right? Look again …

These cells are almost certainly fake Samsung cells. It’s hard to tell, but the first thing to notice is how the lines are not matched monospace – the G doesn’t align directly under the 5, which looks a little bit off. The lot code underneath may have three or four letters from my experience. The text should look like this:

ICR18650-26C
SAMSUNG  SDI
         xxx

If you don’t believe me, just look at these other images of genuine cells from other power banks. It’s also the case for my genuine Samsung cells extracted from an old HP laptop battery.

DSC_7425rp_DSC_6534-1024x412.jpg

DSC_6621

The other characteristic of genuine Samsung cells is the horizontal batch code printing on the metal can itself. You can see it on the cell with the heatshrink removed – but in all my genuine examples, you can actually see at least one set of prints underneath the heatshrink. None are visible in the cells used in this power bank.

If this isn’t enough for you, lets remove the end caps and take a peek there …

DSC_7523

I don’t know about you, but the consistency of the heatshrinking seems quite poor. The two cells even have different cap colours! I’ve always known Samsung cells to have white caps only. The blue cap one doesn’t seem right … why would they pair up two dissimilar cells? If they aren’t well matched, the safety and lifetime of the pack may be compromised.

Look a little more carefully – the spot welds seem pretty averagely done, and the cells themselves have slightly dirty tops. It’s more apparent when looking at the other end of the cells.

DSC_7524

Looking at the underside, it is obvious that there are spot weld tacks left behind. As a result, it is very likely that these cells have been dismantled and removed from existing packs – they might not be new cells at all.

I have heard through the grapevine that there are operations out there which salvage old cells from recycled laptop batteries, remark them and rebuild packs from them. I thought that was ridiculous when I first heard it, but now that the evidence is placed in front of me, it seems quite likely that this is the case somehow.

Lithium-ion cells generally degrade as soon as they are manufactured. Cells which have reached their cycle limit tend to see about 80% of their remaining capacity left, but then they go downhill rapidly and unpredictably from there. I really can’t recommend such cells to end users under any circumstances.

If these are genuine Samsung 2600mAh cells, I’ve never seen any quite this rough and different. The power bank capacity should be 5200mAh, rather than the claimed 4400mAh. I suppose the guys actually know something about this … that’s why it’s not claiming 5200mAh, but then they might just print any number you want.

Performance Test

The performance of this power bank was tested on the new rig, following the same methodology as used in previous power bank tests. The capacity results are as follows:

Load (mA) Run Capacity (mAh)
500 1 3556.703446
500 2 3714.871695
500 3 3673.765736
500 4 3660.550617
500 5 3560.382371
Mean 3633.254773
Range 158.1682496
StDev 71.09503461
Load (mA) Run Capacity (mAh)
1000 1 3355.642847
1000 2 3083.331958
1000 3 3398.524624
1000 4 2902.136176
1000 5 3451.494462
Mean 3238.226013
Range 549.3582864
StDev 235.5389223

At the 500mA rate, the effective capacity was 3633mAh. At 1A, the effective capacity dropped to 3238mAh. It seems the consistency of the charge termination was fairly poor, in more than two runs, the capacity differed quite significantly from other runs. This could have something to do with the cells, or with the performance of the charger IC at different temperatures.

The efficiency of the converter, assuming a 4400mAh capacity, is 83% and 74% at 500mA and 1A respectively, which is acceptably good. But if we use the 5200mAh total capacity as the cell shrink wrap tries to pretend, the efficiency is a lousy 70% and 62% respectively. This is another confirmatory piece of information we can use to conclude that the Samsung cells are likely to be fake.

Lanu-Power-LP-401B-VvS

The voltage profile is on the high side at 500mA, but kept good regulation, which is preferable for overcoming resistance in USB cables. At 1A, the converter seems to have no concept of regulation at all, with the voltage falling below the 4.75v minimum USB voltage requirement before half-way through the discharge. It seems that the output current capacity is a lie as well, and the converter is overloaded. The PCB does shut down correctly once the batteries are depleted, however.

 

RIPPLE-lanu-500ma

The ripple voltage at 500mA loading averages 471mV, which is high – the voltage should sit between 4.75v to 5.25v. With an output of 5.2v already, adding about 0.2v positive spike exceeds this limit, and is thus non-compliant with USB power output requirements. The oscillation frequency is a high 1.052Mhz, which allows for smaller capacitors and inductors to be used, at the expense of losses in switching and inductor core/winding.

RIPPLE-lanu-1a

Increasing the load to 1A increases the ripple to 753.6mV average, which is too high, and again non-compliant with the USB standard.

Conclusion

This power bank is the first power bank where it seems counterfeit cells were used in the construction. The cells themselves seem to have been used before, and may have a questionable lifetime. They are unlikely to have come from Samsung. The conversion circuitry wasn’t particularly special, with regulation difficulties at 1A, and a high level of ripple (but not the worst) which makes it non-compliant to USB specifications and potentially risky for devices.

Posted in Electronics, Power Bank | Tagged , , , , , , | Leave a comment