Haswell Could Have Compatibility Problems With Older PSUs

Status
Not open for further replies.
Good. This will hopefully get people to throw their fire hazards (that is how I officially refer to cheap power supplies when talking to customers).
 
It will be interesting to see how off-the-shelf manufacturers handle this, especially in inexpensive machines. They will need to start putting quality PSUs even in the low-end machines. I am sure the big boys like Dell and HP will be fine, but what about E-Machines and the such?
 


PSU oems will just start making cheap units that support the new CPUs. There will be enough demand to justify it.
 
LOL, in comes CUSTOM WATER COOLING. throw in a pump, and one or a few high powered GPUs, and you have a "high enough" powerdraw at idle 😛. of course, all those people should not be running crappy GPUs
 
Most cheap PSUs already have internal loads (resistors) to guarantee the minimum load they need to remain reasonably stable.
Where Haswell may be more of a challenge to cheap PSUs is response time to large load swings when the CPU keeps bouncing between C7 and full-power: too little feedback bandwidth means the voltage rails will sag when the CPU wakes up which may trip under-voltage lock-outs or surge when the CPU goes to sleep and trip over-voltage lock-outs. Depending on how fast Haswell can switch from full-power to sleep, the PSU's feedback delay could potentially encounter situations where the CPU is already going back to sleep while the PSU is still ramping up from the previous wake-up or vice-versa.
 

Why so interesting? 0W means 0A at whatever voltages. It's like removing the Processor from the socket and measure the voltage being delivered at the socket pins.
 

Nothing special/interesting about it: a Lithium battery delivers ~3.6V/cell open-circuit voltage, a fresh alkaline battery delivers ~1.65V/cell, a lead-acid cell delivers ~2.35V, etc. Mains voltage is 115-120V regardless of whether or not you have anything other than your multimeter plugged in. Your PSU's 12V rails are expected to hold 12V regardless of whether or not you have anything plugged in, same for 3.3V and 5V outputs.

Voltages can exist without a load. Voltages and electrical fields are everywhere; even in the vacuum of space.

Electrical current (measured in Amperes) on the other hand does require some form of conductor and load to exist.
 
I hope they don't come up with the "Your mouse sensor system is obsolete when using Haswell CPUs. We recommend you upgrade your mouse to an approved model to fully experience what Intel has to offer"... 🙁
 
If your PSU fails at zero load, you're doint it wrong. Every decent power supply should be able to pass the zero load test, where it operates with the pins shorted to turn it on, and nothing plugged in as load.
 
math1337 - that zero load test statement is inaccurate. A modern advanced PSU may or may not activate depending on voltage protection circuitry. Some will require a simple load, even a fan, to power on and satisfy safety protections.
 

It isn't so much about "satisfying safety protections" as it is providing sufficient drainage for the energy stored in the HF transformer's leakage inductance which keeps dumping energy in output capacitors, causing output voltages to drift up until over-voltage protections trip.
 
The classic problem of switching power, at very low current, it cannot regulate the voltage properly, a simple solution for this is place a dummy load on one of the connectors to achieve the minimum current for the regulator to run properly
 
I discovered that if you comment on article do not refresh the page to check your comment. Either close the tab or click out of the page because each refresh causes a duplicate comment.
 
Status
Not open for further replies.