i have one word for all of you: paranoia.

has any one has a case of this being a problem?

me thinks the answer is a big fat NO.

CatDog2,
As u might know, I've put up a page with info on this with info taken from this thread.
Also just have added your latest testresults.

You find it all here: http://xbox.bula.nu/resistor_install_1.6.asp

Would be nice if you could take the time to check trough it so that any info stated there are correct and up to date. Easy to miss out on new stuff in this rather large thread..

Anyhow it also would be nice if you would drop a mail or PM when there's any new findings about this issue, so that I could update the site w that info.

I have an v1.6 xbox and i have installed the x3 i got yesterday, So i need to solder the resistor and cut the wire? I want to use it with the mod chip either on or off, would this affect live as i am using live when the chip is off(going to)? If i dont install the resistor how much time would it be before the xbox died a couple of months? Sry if n00bish, just trying to be on the safe side.

I have read on many Xbox forums of people quoting – “My console worked fine for weeks and then Fragged for no reason , please help”

Ok we don’t believe all we read but because of all the commotion about this issue something had to be done.
We opened a 1.6 and 1.6b Xbox and did some tests.
Our findings confirmed what has been written with currents of over 65ma recorded on both consoles. Although nothing has been proved or disproved we believe that this current is too high. With careful resistor selection we were able to get the Lframe line down below 4ma with no problems running in any bios mode. If this is the case we could be overdriving the MCPX chip by over 12 times. The chip may well be able to cope but for how long, weeks, months or years it’s a bit of a lottery.
What can be said is that one manufacturer has already acknowledged this problem and produced a modification. You can do it in half an hour at almost zero cost.
Both our boxes and now converted and working great.

We believe this to be a justified modification…

I've got a 1.6, and I think I'm going to do this, just in case. If it turns out to not be a problem, well, I haven't hurt anything by doing it.
Is this the surface mount resistor reccomended?
http://www.digikey.com/scripts/DkSearch/dk...w=87430&Site=US
Does anyone have one they could send me, as DigiKey has a minimum of 10, and a $5 charge for a small order.

-Duck!

well i just talked to my dad (who is a fully trained technician and works on this all the time). he said that the chip can no way supply 65ma of current, so anyone who gets that reading is doing somethign wrong/cant use a multimeter.
and also, with chips, they eitehr work or they dont. if was goign to fail, it would blow up straight away, or get hot and blow (but the longest time that that could be would be like 10mins).

so this problem does not exist, and there is no reason at all to have to install a resistor or anything

Duck of Death - Use a standard 1/4 Watt resistor. This however is difficult to solder to the PCB (leg of component to big) We used some rigid copper wire the size of the PCB hole and soldered that on. We the soldered the resistor to that thinner gauge wire.
Make sure you clean the PCB pad to remove the resin or nothing will adhere.

Ferrari rulz 02 - We suggest you dad conducts some test of his own before stating any facts. We are not saying it will blow any chips. Just confirming the fact that overloading certain components may cause them to fail at some later stage. Nothing last forever but anything we can do to extend an electrical products life must be a bonus.

1) Assuming the current is >65mA, then your dad is agreeing that this level of current is extreamly high - much higher than the ratings of similar* components he is used to working with.

*by 'similar' I mean cmos components with complexity near that of the MCPX, as datasheets are obviously not available to the public.

2) I suggest that your "trained technician" dad take some measurements himself. 65mA is equal to an equivilant series resistance of approx 50 Ohm, which is not outside the capabilities of some ICs. The measurements have been done by a number of independant people, including markone, and hippo here (which also suggests that the Duox2 has this issue).

3) As an example, static damage often does not cause a device to fail immediately - it might be weeks, months, years later on it will fail. Your dad should know this. This does not mean that static damage is not a problem - under a high power microscope it definately is. This is a similar long term effect.

I have never stated that the high current will "kill your xbox", but it is almost certainly over the maximum ratings of the device and hence this is a warning about a possible problem which can easily be prevented

the duox 2 finaly has the problem?
please catdog make the test and answer!

In the data sheet for these CPLD's we're using it pretty much tells you straight up 'do not ground out more than 64ma through one pal block of the chip'.

On CMA32 there's D0 and D1 so you can use 2 pal blocks for grounding. Most Aladdins have D0 and X although some cloners never read the data sheet and just connect X and D0 to the same pin and thus the same pal block - not too bright. On Sunday chip there is D0 and D1 aswell but with the new life support signal you'll never use them. They are at least on seperate pal blocks though.

Well that is absolute nonsense. That is a fact. I am a fully qulaified Technician. this is all I do for a living and the statement above is rubbish. Take for example a BU2508D as Line out transistor in a TV set. I get Sets in all the time with this part short circuit. the cause quite often is the line drive frequency is wrong which causes the Transistor to overheat as it is drawing too much current. A repeat failure of the transistor if the cause is left unresolved can be anything from a couple of minutes to over a week sometimes more so to state excess current would cause failure in at most 10 mins is not accurate at all.

What is this statement. Current is not SUPPLIED. The current is drawn The current in a circuit supplied by a fixed frequency output is dependant on the design and discrete components in the circuit.
Take a simple 12v power adapter that is rated at 800ma. This does not mean its output is 800ma. The current drawn from the adapter is dependant on the device connected to it. If its connected to say a programmer whose total current drawn id 8ma then that is it 8ma. The adapter does not kick out 800ma at the programmer.
If the chip determined the actual supply current which it does not there would be no problem because the current supplied could never exceed the output from the chip.
The theory behind all this as it came about was if the Lframe was held permanently to ground what would happen. So you measure the current in series with the circuit in both conditions (chip on chip off). Theory is the circuit draws excess current if the Lframe is held to ground permanent. To reduce the amount of current draw the signal can be pulsed. Some chips supposedly do this.
catdog2 was just looking at a way to reduce the current drawn when he posted this thread. Less current drawn = less thermal stress = better for the circuit concerned.

hey chancer

Have you by any chance measured this with the new X3 CE??

looks like the chip layout is quite different.. wonder if Xecuter made some hidden changes

if not i will at least measure the D0 on a V1.6 with the X3 CE.

I assume it will be held low as well.. but you never know.

jROC

Update 18.00 GMT
Right just come from the workshop now after running some tests. The most interesting being without a modchip running and measuring the current in series. the current is 68ma. thats with the box running normal. I checked it 3 times to be sure.
With a Duo X2 in and enabled the current drawn is 52ma. This is less than the current drawn bog standard without the chip. Is this a problem ? No I don't think so.
Not able to test with an aladdin on this as I have none left in stock.
The meter is a Fluke 77 calibrated in Novemeber 2004 and used daily in repair work on TV Digisat etc so is not some cheap Radio shack one.
Any thoughts or conclusions from anyone

Not had chance to measure an X3 yet. Its a chip I rarely buy as all my customers are poor so they always jump for the cheapest price install. I would be interested if anyone could measure the current before and after and post the results

Ok..i'll do a bit of testing on teh X3 CE.. I already know the original X3 holds it low all the time. I measured it.

And I suspect the X3 CE does as well.

I agree with you tho.. this might be a potential issue over the long term.
But if its like 5 years.. then i would say no worries.
Cause the console will be outdated by then.

I've done alot of V1.6 with all kinds of chips..so far nothing returned cause the chip fried the box.

jROC

I would be curious to know this as well

Okay, i have a 1.6 Crystal with SmartXX V2 inside. Should i get worried?

Yes, I'd be worried, but don't be. Smartxx has noticed this problem, thats why they created the 1.6c rebuild board. It was just recently released (annonced, January 17,2005) so you prolly don't have it installed. So do as they suggested so far (install the resistor & cut the trace), OR pull your rebuild kit/LPC wire rebuild and install the new one shown below.

Chancer are you saying that this is not a problem then, because it was stated earlier that the current draw from a x2 or 3 was 65 mA and if you messured 68 mA with no modchip installed then we should not even be having a discusssion about xbox failures!

I'm lost I'm going to have to do these tests myself now to see what really is happening.

How did you messure in series with the chip did you cut the trace?

Thanks

Kbarton

The picture at the first page doesn't work for me, so is there somewhere a pic which informs which trace i should cut and where i should install the resistor?

Can you please clarify exactly where this measurement was made between. The current that flows without a modchip is virtually zero (current will only flow on data transistions, and will be in opposite directions - no DC current flows). I have checked this by cutting the LFRAME trace and putting a multimeter in series with the cut - maximum measurement values shown were 0.01mA (which is the resolution of this meter). An oscilloscope with a current probe is required to make any further measurements that provide useful information - but the values are so small they are not worth measuring.

Your measurement suggests you placed one lead on LFRAME and the other lead on GND? If so then this is the same as grounding LFRAME directly (through the low resistance of the meter). If this is not how you measured this value, can you please provide a detailed explanation or diagram of what the measurement is of.

Thanks

Mika-- The diagram can be found here:
http://img.photobucket.com/albums/v238/cat...storinstall.jpg

Sorry I have confused everybody with my poor explanations and bits missed. What I did was cut the trace and measure with the meter either end with the chip off and the current was 0.005ma which I would assume is normal as there is no load nothing on the output.
What has confused me is when measuring from Lframe to ground (which I realise will increase the current flow due to the internal resistance of the meter) I though the figure of 68ma was very high. The point i was trying to make or clear up is why should the current when measured the same way with the duo X2 enabled be lower than with the chip disabled if the Duo was now thought to ground the Lframe direct which surely would result in an increase in current drawn over the amount caused by just having the meter connected this way. Surely this would indicate that the Duo does in some way pulse the Lframe. Or is there still a problem?
I know it would have been better to measure the chip enabled mode with the meter in series with the D0 pad but I could not get the chip to enable.
I hope this makes more sense, probably not.

Well, thats a little hard to do cause my SmartXX LPC-rebuild board is over those points, and im not good at soldering so little components...

So, i will do nothing, hopefully my box doesnt blow up

I finally got hold of a duox2 chip and tested it:



The current measured was 48mA, which is lower than some other chips but is still higher than it should be.

The duox2 does not pull LFRAME to 0V (the modchip cannot sink this level of current?) but only down to about 1.2V. This is still seen as a logic low by the xbox so has the same effect as grounding it completely (this is not good practice - the voltage should be as close to 0V as possible in a CMOS circuit).

The confusion about Duox2 pulsing LFRAME has come about because of this. Shown on the diagram below is another view of LFRAME with the modchip enabled. Without looking at the scale it looks like the line is changing between low and high levels. Notice that because it is never going up to the arrow marked at 3.3V it is actually changing between low and low (as far as the xbox is concerned). The X3 modchip operates in a very similar manner (pulls level down to about 0.8V) and has a similar current to the duox2 (51mA measured on X3).



Conclusion: DUOX2 does not pulse LFRAME. It holds LFRAME logic low in the same way other modchips tested do.

My head hurts:(




Can we dumb it down to the level of about a third grader:p


I think most of us just want to know what chips might cause premature failure and which ones wont.

All of the chips I have tested might cause premature failure:

Aladdin Advance
Aladdin XT Lite (4032)
Aladdin XT Pro (4064)
Duox2
Xecuter3 (I would assume X2 is exactly the same)
Xenium Ice

The smartxx can be used with the "smartxx 1.6c rebuild board" to avoid this issue (but without that rebuild board it is also a problem).


Sorry about the level of the description, it is given to justify any claims made for those who understand it (rather than just making irrational claims with no supporting evidence).

too bad, I just installed a DuoX 2 when I read that it also has this issue...
So I went right ahead and cutted the Lframe of my box to make it an always on method.
but what do I do with the BT connection?
I now have it the way it should be connected when installing the duox in on/off mode, but my eject button does not realy work like it should.
when I touch the eject button, the box powers on, powers off, powers on and boots with chip (wich is pretty much all the box can do, considering I cutted the Lframe forcing it to boot to LPC right?).

anyone know of an other point to solder the BT to? I tried ground already, but that doesnt work, the box won't boot then (frags).

ps, I already removed the D0 wire to the modchip ofcourse

Connect BT on the Modchip to Pin 9 on the Modchip (should be the second Pin down from D0 connection)

ModdingManiac, thnx for the solution
that indeed did the trick, it suplies 3,3 Volts to BT the same as it did when pressed the power button I figured out.

well, bmmods has the 1.6c smartxx rebuild boards in stock now, so maybe someone wants to test one out..........

I think we have modded over 500 xboxes with all kinds of chips and none of the buyers have reported of any problems yet. We started to mod these boxes as soon as they were in the stores. Compared to the laser problem on pstwo v12, I think this problem is of no concern. Doesn't problems like this usually have both long and short term consequenses?

Regards,
Digistore

I'm starting to regret that I bough executer chip.. a thread asking about the issue was quietly deleted from their forums.

Conclusions from that are that it is an issue and that they don't bother to fix it because by the time the xboxes start to blow up they have already cashed out.

Thanks!!! That's the kind of info I needed:D


For the smartxx 1.6c rebuild board I assume you mean this:
http://www.divineo.com/cgi-bin/div-us/xb-sxx-sle


If I could understand what you were saying in your other posts I'm sure I would appreciate the info. So please keep posting your findings and just remember to dumb it down every once in a while:p

Nothing, i was wrong.

X3 Config Live Build 2666
[1959 -> 2666]
* [Fixed] Corrected LFRAME / D0 pull to 0V timing so it's no longer constant

Yep, just saw that.. looks like I was wrong

Just have to wonder why they didn't reply with that instead of deleting the thread, it would have ended the FUD on the spot.

Interesting.

My 1.6a behaves exactly as you described but with a Xecuter 2.6ce chip installed. I too have cut the Lframe trace on my box (but have not gotten around to putting the resistor in yet). I wonder what I need to do in order to prevent this from happening on my box. Can't do the solution suggested a couple of posts above as I don't have a duox chip.

On a related (maybe) issue, my IGR also causes the box to reboot. I wonder if the solution to the eject (reboot problem) will fix my IGR reboot?

Anyone else observed this with their X3 or X2.6ce on a 1.6a? I'm running M8_16.bin (not the plus).

WEll thats wonderfull that they did that for the x3, but what about all the rest of us that use the X2.6chips. IS there as similar upgrade for the x2 bios in the near future???? ...one can only hope right?

that wouldn't apply to just smartxx, though, right? any modchip mounted on that smartxx rebuild board (like xenium ice) would no longer have this issue?

thats nice. i thought it wasnt a problem lol. well, anywho, i hope they get it out soon so people can test it.

I thought this was hardware issue, how would a new bios/OS help?...If thats the case then why wouldn't all other teams have come up with a fix already, including Evox?

someone pwned the pic host!

anyone have a mirror?

I just mentioned that on xecuter forums. I said I find it intresting how a problem thats not a problem is fixed. This can be fixed with software cause tx can recode the eeprom on there chip to fix this. Not to be mistaken for a bios...The world will tell when this is released.

When's this bios slated for release? Anyone know if they will make a 2.6 one?

A bios has no chance of fixing the D0 problem. This is called lying and it's exactly what X3 team does best... wait no that's price gouging they do best then right behind that is lying. Keep it straight.

It's possible that the bios could update the CPLD and if I was lying and saying that we were fixing the CPLD with a bios update that is the method that might pass some scruntiny. They on the other hand decide to lie about going into the CPLD code via the I2C bus and it's a funny joke. You should tell them X3 guys that if they make up a whooper they should try to use some facts or otherwise try to make some sense.

Pretty much keep in mind anyone releasing a bios that updated their CPLD code would not be able to release the bios public since the end users would basicly have the CPLD JED file and could easily make their own clone.

The xecuter bioses; are encrypted so they could include the code to fix the problem. And you know if it doesnt fix the problem it will be on the bug report and peopel will bitch In this case I dont think they would announce it if it wasnt true there x3 bios is their. In team x's they never said that the lframe low didnt affect there chip they said that the held low lframe wasnt anythign to worry about...much like smartx they did address the problem after multiple requests and soon enough we will see if it fixes the problem. And there chips are not expensive they are middle fo the road...smarttxx chips are by far the most expensive. Followed by the x3 then the xenium. for the quality products and stuff that x and smartt x delievers ill pay the xtra. thank you.

Why I have to use an EVOX bios if I have an xecuter 2.6 chip??.....xecuter team...sucks..big time..

In either case.. i doubt that the X2 will get this fix.

WHy??

Well think about it.. it uses an actual switch to control the D0.

So so matter what bios you put on there.. its always gona be the same

Whether the D0/Lframe problem is fixed with the X3 remains to be seen... and until the new bios for the X3 is released.. we won't know.

I think it would be possible for them to fix this with the new X3 bios update.
not 100% tho

jROC

So you are from the X3 team and you are saying that the bios update will flash the CPLD and fix the D0 signal on the CPLD? First off it will be so long before they ACTUALLY release their bios you will have forgotten about the features they promised.

The fact the bios is encrypted means nothing quit talking nonsense.

Im not from any team. Infact I own chips from xenium, and smartxx. And the bios will be realeased shortly. Its in the last stages of beta testing. Im not worried about it to be honest. And no I actully wont forget about the features they promised. you sit here and rip on xecuter you should rip on the other teams as well namely teamxodus. They both do fucked up stuff. But the fact remains that xecuter can fix this problem via a software update. NO OTHER CHIP MANUFACTURER CAN. And the fact that the bios is encrypted does mean something so. Now rather it will fix it or not I dont know but being its a touchy issue. Im sure it will be. As usuall there will be 2-3 versions of the x3 bios 1.0-1.5 then 1.6/1.6b