Extend The Life Expectance Of Your Modded 1.6
Posted 10 January 2005 - 08:15 PM
THERE IS NO PROBLEM, AND A RESISTOR IS NOT NEEDED. It is an extra step that you are taking, that doesn't help anything.
Look, I am an electronics engineer, retired from the US Air Force, where I worked with sensitive aviation weaponry circuitry for the last 20 years. You can not test the LFRAME circuit with a multimeter or a general-purpose oscilloscope. You need logic analyzers and a sampling scope, something around 2GHz and at least 8GS/s.
What I can tell you is that the LFRAME circuit is based on an OPEN COLLECTOR design (Google that term if you don't know what it means) and therefore simply the presence or absence of voltage means nothing. Not to mention the fact that the motherboard already has a 10k pullup resistor on that circuit...
Geeze. This whole thing started by someone's speculation, not on actual facts. Have ANY motherboards actually shown up that have been damaged by this yet? No... and there won't be any either. All of you that are hacking up your motherboard traces and installing resistors are just following a paranoid suspicion based on speculation rather than conclusive facts revealed by a proper scientific approach by properly trained personnel. You are more likely to damage your console yourself rather than the subject of this topic...
Posted 10 January 2005 - 11:28 PM
Yes I do know what an open collector design is and I don't need google to tell me but I don't know any more than you do if that type of design is employed. If you can deduce the circuit layout fo a multilayer PCB by looking at it then you truly are a superman.
You don't know this either anymore than anyone else does!!
Before you ask yes I do know about electronics I have a C&G 224 and an HNC in electronics as well as a formal 5 year apprenticeship and 28 years experience in the trade.
State facts of how you KNOW the design of the circuit and I may take more notice of you because at the moment you are guessing just like the rest of us.
Posted 11 January 2005 - 03:44 AM
Posted 11 January 2005 - 07:22 AM
While the xbox is booting when the LPC bus is active then yes a standard multimeter is not capable of providing a meaningful reading. However, after booting when the bus is idle a multimeter will provide a good indication of the voltage and current.
-a 10k pull up resistor on a 33MHz data line?? Even the smallest amount of capacitance would introduce problems with the risetime of the digital signal.
-the maximum current that could be drawn would be 3.3V/10k = 0.33mA. This is not the case - the current (when the bus is idle) is over 60mA.
Measuring the resistance between Vcc and LFRAME results in 10k so no other resistors are being used to pull up the line. Cutting the LFRAME trace before the pull up resistor and measuring the currents results in:
LFRAME (from MCPX) to GND -> 61.5mA
LFRAME (from xyclops and pullup resistor) to GND -> 0.33mA
this clearly shows an active driver is being used to drive the line.
You have no idea of my level of training and equipment used, so how can you claim that I am not "properly trained personnel"? I don't have access to the xbox schematics/datasheets - but very few people do and this hasn't stopped xbox mod development. The fact remains that there is 60mA being drawn from a single 3.3V digital line. The issue is whether this is high enough to cause long term damage or not, to which I can only provide an educated guess by comparing to maximum ratings of other common devices.
Regarding motherboards dying, plenty of v1.0 motherboards have died due to the MCPX failing. Drawing more current from this can only increase the already existing failure rate - wait for a year or so and then we will see how many motherboards have died. I have already stated that this problem is "long term" - read the title of the thread. I dont anticipate instant failure. I, along with many others, don't want to have to buy a new xbox every year due to a problem which can be prevented.
Yes cutting the trace on the motherboard is dangerous and can cause more problems than it solves when done without care. But this provides an option for those who want to do it, and it works well when done carefully. As with all xbox mods, it is done at the users own risk and can damage the xbox. Ideally the modchip developers would tristate the LFRAME line when it is not in use and there would be no problem at all - note that hippo has already shown this can successfully be done.
Posted 13 January 2005 - 10:50 AM
Posted 15 January 2005 - 11:38 PM
Posted 16 January 2005 - 11:12 PM
by the way, for any one new, cutting the trace is the easiest bit and requires no real knowledge of electronics.... and i shake like a bloody leaf mate. bump.
Posted 17 January 2005 - 02:16 AM
Still, a very interesting tread.
Posted 17 January 2005 - 03:23 AM
Still, a very interesting tread.
The X3 holds the LFRAME signal low despite what they state on the xecuter page. The Xenium ICE does this as well - they also claimed that it was pulsed low. I will provide screenshots tonight or tomorrow to show this.
Posted 17 January 2005 - 04:43 AM
Posted 17 January 2005 - 07:39 AM
First up, this is what the LFRAME signal looks like with no modchip installed.
A zoomed in view with the 33MHz clock shown as a reference:
A zoomed out view with one of the data lines as a reference:
What this shows is that the LFRAME line is normally high (3.3V) and is only pulsed low when it is in use (during the short time that the bios is loading). After the xbox has booted it is in a high state.
I fully installed an X3, used the alternate 5V point, and installed the DVD/harddrive so the xbox was fully working. I used the latest bios version available (1959). Here are the screenshots:
This shows that the X3 holds LFRAME low, and it is not pulsed as they have stated. Note that in this case 'low' is ~0.8V and not 0.0V
I measured the current flow once the xbox had booted: 51mA
Next up was the Xenium ICE, again with the DVD/harddrive installed so that the xbox was fully working. After quite a battle to get the OS updated to the latest version (I ended up putting it in a non 1.6 xbox before it would update??), I finally captured the following screenshots of the OS and also loading a bios (M8_16):
Again LFRAME is held low rather than being pulsed as they have led us to believe. The xenium ICE pulls the voltage very close to 0.0V, and therefore generates a larger current flowing from the MCPX.
This LFRAME problem is an issue for the X3 and Xenium ICE
The information provided here (as it was from the beginning) is to provide awareness to a possible problem and one possible solution. Testing on these two chips was done due to the number of requests and confusion about them. It is not here to defame any particular modchip teams (as far as I am concerned they achieved that on their own with the manner they responded to the topic). Smartxx have released a product which apparently deals with this issue, so I have not tested it here. Aladdin chips do hold the line low and since this hasn't been disputed I havn't seen the need to spend time showing it.
As stated many times throughout the thread CUTTING THE TRACE IS DONE AT YOUR OWN RISK and can damage the motherboard if not done carefully.
Posted 17 January 2005 - 07:46 AM
On v1.0-1.5 xboxs the current is only very brief (much less than 1/10 of a second) rather than continuously flowing, so it is not a problem.
(Also it is from the TSOP rather than the MCPX - so even if it could be a problem a modchip would still allow the xbox to work if the TSOP was damaged).
Posted 17 January 2005 - 11:07 AM
0 user(s) are reading this topic
0 members, 0 guests, 0 anonymous users