MUST READ : LG4167B and Power Supply Issues

About a week ago I bought a 4167B and had a few minor problems with it. Essentially I found it to be a fairly poor reader, with frequent of ECC error's and "damaged blocks" when testing with CDspeed, even when the same disc scanned perfectly in my other burners!

After I placed this burner in my older computer all these issues miraculously disappeared. At first I stupidly thought that the drive had just loosened up and "run-in" and that's why it was now working good. But when I returned this 4167B to the original computer all those previous problems reappeared. Only then did the thought occur that it was probably a PSU issue. (BTW I've never had any instability or other PSU related problems with this computer before so I really wasn't expecting this.) You can read about this whole saga HERE

Anyway I had a nice new PSU here just waiting to be dropped in, but I thought that before I did that I'd make a few little tests and try to determine if it was really the voltage levels or actually the voltage stability (ripple) that was causing the problems with this burner.

I made a really simple test where I just placed a pair of large capacitors on an adjacent molex connector to see if this made any difference (see pic below). Interestingly this cured the problem immediately and completely, so it's voltage stability that's the culprit.



After doing a round of tests and being quite certain that all problems were truly gone I then decided to remove the capacitors one at a time, first from the +12 - 0 volt side and then from the +5 - 0 volt side, to see which power rail was really causing the problem. I removed the capacitor from the +12 rail and the problem did not return. So I have concluded that it is stability on the +5 volt rail to which the 4167B is overly sensitive. I've left the capacitor on the +5 rail in place for now and might run it for a week or so like this to be sure that it's fixed the problem, it sure seems to be fixed.

Very good test uart, many people underestimate the importance of the power supply to the overall stability of their systems.
I bought a 4167B today and connected it through USB2, it will be interesting to see how it performs...

Does the USB enclosure have a built-in PSU or is it supplied from an external "power brick"? I'll be interested to here how it goes because this drive sure does seem very sensitive to power supply quality.

BTW, Most of the other posts regarding problems that were cured by replacing the PSU were cases of the drive behaving very badly, like not even recognizing blank DVD's at all. But my case was much more subtle, mine always recognized my blank media and always burnt it quite well. It was only in read-back that it was slightly marginal and even then it was only with some media types and not others. So it would have been very easy to dismiss my problem as a slightly below standard drive or alternatively to just blame it on the media. This indicates to me that this power supply quality issue is a continuum that can vary from only very minor degradation of read performance all the way up to the point where the drive is totally unusable.

Good test. Obviously LG is paying insufficient attention to filtering the input power in their design. Has anyone looked inside the case to see what sort of capacitor they have across the power input?

Hmmm, it occurs to me that these 4167B models are being made in a factory in China for the first time with LG. The mandate is obviously to reduce manufacturing costs, and one of the ways you save money is by using cheap components - like capacitors. Capacitors are the most likely electronic component to fail or be defective, and the effect is quite often subtle - like failing to perform the proper filtering function. Is it possible that there's a bad batch of power-filtering capacitors in the 4167Bs, and therefore only some people are affected?

Yep, the LG burners suck many more power compared to other dvd burners.

So never underestimate the good vs. bad PSU problem.

Hopefully this can help other users to solve the riddles...

[QUOTE=uart]Does the USB enclosure have a built-in PSU or is it supplied from an external "power brick"? I'll be interested to here how it goes because this drive sure does seem very sensitive to power supply quality.

It's not an enclosure, it's a PATA to USB2 plug-in adaptor with its own (quite small) power supply. I had another one of these and its power supply was huge, brick-like. This is one of the newest models with an ON/OFF switch on the adaptor, a power led and a much smaller and lighter power supply. I connected it via IDE first and booted into safe mode to upgrade the firmware to DL13. Then plugged it on USB2 and burned three PRODISC-R04-04 x16 DVD+R with very good results. Looks like my USB2 adaptor's power supply is decent enough

Thanks again uart for pointing out the power-related issues of this drive.

My scans posted here:
http://club.cdfreaks.com/showthread.php?t=156313&page=5

Chef the 4167b is rated at 1.5A on the +5 rail, this is similar to most other burners. The truth is that this burner runs no hotter than what's typical of most other burners so I really don't thinks it's a straight question of power consumption.

In my opinion Ames200 is on the right track, I also believe that this issue is a sign that LG has made this model too cheaply and probably skimped on some important PSU filter/bypass components. Every other model/brand of burner I've tested works fine in my computer, even if I have two other burners connected and and running at the same time (on the fly copy) everything is fine. Even if I overclock my computer from XP2600 to XP3200 equiv and have two other burners running then it's still fine and has no stability problems (I don't normally overclock BTW, that was just a test to see if my PSU could take it). But then I place just this one optical drive (the 4167b) in my computer, no overclock, no usb peripherals or other power consumers connected and the thing can't work properly due to lack of power. I really think this is a problem with this drives design, especially after seeing so many others here with similar issues.

Interesting info, uart.

Thanks!

It would seem to me that the problem with the power supply is excessive ripple and possibly noise on the 5 volt line upsetting the logic chips on the drives main board. For hf noise try also a ceramic capacitor across the electrolytic.
You are doing what the manufacturer of the power supply should have done and fitted good quality electrolytic capacitors. But there is another possibility.
If the power supply is old it could just be that the capacitors have "dried out" that makes them lower in value and unable to do the job, particularly if the cheapskates don't build in any slack and use the minimum value to keep the costs down.
This increases ripple on the power and they can't deliver sudden power surges required causing the voltage to drop and hence problems. Some drives are better designed to overcome these possible shortcomings others are not so lucky!

Very good info uart. One thing I'd like to point out though : the 4167B isn't the first writer LG has had made in their factory in China. Later models of the 4163B were also made there (earlier ones were still made in Korea). You'd think they would have worked out the kinks in drive manufacturing by now having had one model to work through in that factory. Could be just a case of trying to cut costs too much.

The problem is that this has made me think about other drive problems people have had which could not be explained. I've started wondering how many other drive problems with other writers could be related to this.

Yes that's a good point, smaller (lower value) capacitors generally have lower parasitic inductance than larger ones and thus can be more effective at high frequencies despite there lower capacitance value. This advantage can be diminisihed if the cacpacitor is not placed nice and close to the source of the problem. If the extra capacitance was designed into the drive (by LG) and placed at just the right location then much smaller capacitors would probably be more effective than the great big one that I used externally. Still a few 0.1uF ceramics in parallel with an electrolytic would probably have been ideal but unfortunately I didn't have a very large selection of capacitors here when I did the test.

Yes without more exacting tests on the actual voltage level and ripple, both with and without the extra filtering, it is not really possible to say for certain that this is LG's fault for under-designing their power filtering on this drive or whether it is the fault of my PSU, that much is definitely true. But considering that none of the other burners that I've tested (and it's been quite a few) have suffered any problems then you'd have to lean toward this being an under-design problem from LG.

Actually this type of issue of exactly who's fault is it when two things don't work properly together (in this case my PSU and this burner) occurs all the time with PC's. Take the example of say a USB device which works on one computers usb ports but not on another's, do you blame the USB port of do you blame the USB device? Some will point out the the USB port works with other devices but just not this particular one, so it must be the devices fault. Others will point out that device has been tested and works on several other computers but just not this particular computer, so it must computers USB ports fault. Who's right? In practice it often turns out that both the port and the device are marginal and both operate right on the very limits of (or slightly worse than) their allowable specifications. If either one of the two had any safety margin built in and operated slightly better than it's worst case specification then it would compensate for the other's shortcoming and they'd work together. It's alway hard to say exactly which one is at fault in this type of situation.

Yes, considering how incredibly inexpensive DVD burners currently are there must be a constant pressure on them to keep the manufacture costs down. And high quality capacitors are often a relatively expensive component. Still they've done a good job with the drive otherwise, it's a great little performer provided that your PSU is good. I'd encourage anyone else having issues with this burner to have a go at adding some capacitance across the molex before forking out for a new PSU, particularly if your system is otherwise fully stable with no other signs of PSU weakness.

PS. Not only does my burner now read media much more reliably (which was the original issue) since I added this capacitor but I think my burns have actually improved as well. Note that the burn quality was still acceptable before the mod but I think it's slightly better now. With my current batch of GSC003's I was getting scans with about 400 PIF's (total) when scanned on my liteon. Last night I burnt my first disc since doing the mod and it was only 102 PIF's total for a full disc. There's always a bit of variation in PIF counts from one disc to another so I'm reluctant to conclude too much from just this one, but it looks promising.

---

Last edited by uart; 11-02-2006 at 18:50.

Hi:

Can you post in detail how the capacitors (and what size) should be placed on the molex connectors?

Thanks

The one shown in the picture is rated at 10 Volts and 3300uF, but that might be a bit of an overkill. No one has done enough testing to know for certain what the best value/type capacitor is but I'd suggest an electrolytic type of value somewhere from 100 to 1000 uF in parallel with a 0.1 to 1.0 uF ceramic type.

The capacitors should be connected between the red and black wires somewhere close to the back of the molex power connector. Electrolitic capacitors are polarized so you must be careful to connect the terminal mark negative "- - - -" to the black wire (and the other terminal to the red wire).

If you do decide to do it just be very careful that you dont short circuit anything!

What type of PSU do u have mine is a ATX 450 PSU. The 4167B was wokring before i did about 12 dvd burns and then now it doenst even read any BLANK DVD-R media. I only have this drive(4167B) and my Harddrive on my computer, no other cd roms-dvdroms etc. or floppy drives are on this computer.
What if i used another IDE cable??. Currently i Have the harddrive and this 4167B on the same one. What if i put in another IDE cable and had 1 cable going to thehard drive and another going to the 4167B???

I agree with you 100% but people trying/doing this must take care. If they are not used to using a soldering iron and general circuitry they should get a friend who knows what they are doing to do it for them.
I would suggest a value of 470 to 1000uF with a disc ceramic of 0.1uF or there about will do the job.
If you are not happy modifying the wiring harness from the power supply then I suggest that you purchase an extension coupler or two way adapters some shops sell for powering up two drives and modify that outside the computer and when you're happy with it just plug it in.

I believe it is best practice to have the drives on separate IDE channels; the HDD on channel 0 and the DVDRW on channel 1 both set to master.
Use good quality 80 wire IDE connectors and you automatically avoid problems on your set up.
Many people will have already done this but others who have bought a well designed motherboard; power supply etc will wonder what all the fuss is about.
I have a 350watt power unit 2 HDD's an LG 4136B and Benq 1650 as well as sound card, extra USB card modem and a G force MX480 card and I do not have these problems even if I copy on the fly!

How owuld i put them on sepearte channels? Just by putting the HDD on one IDE cable and the LG4167b on another IDE cable corect?

Yes, correct. Putting source and target drive on a seperate IDE port/bus/channel/cable or how ever you want to name it is neccessary for high speed burns.

Heyyy i tried puttin the HDD and the 4167B on 2 sepearte IDE cables and still the problem occurs. The LG4167B is styll unable to recgonize blank dvd-r media

Is it just blank media that causes the problem or do you have problems with pre recorded media?
Does the drive recognise any DVD’s?

Thanks for the info. I probably won't do it right away because I don't have problems with it recognizing media. But I do keep it in a USB2/FW enclosure. Something to keep in mind if one day it begins to flake out more than usual.

Heyyy WEEDOUGIE it doenst recognize BLANK media howvere it does read DVD movies and blank cds.

As you are also exploring this fault on another thread with more information than you are giving us here on this thread, I suggest that you confine your enquiries there.
Best regards.

---

Last edited by weedougie; 14-02-2006 at 15:41. Reason: Change text.

The capacitor is connected in parallel to the DVD drive, is that correct?