I have a 2009 Mac Pro 4,1 (firmware upgraded to 5,1) that I'm trying to install a pair of Xeon X5690 3.46Ghz CPUs in, but when I power it on, I don't get a boot chime or anything displayed on my monitor. I tried playing around with the amount of RAM, but only change I got was the Mac would start switching off immediately. I tested the X5690s in a 5,1 machine and they worked fine and the Mac I'm attempting to install them in works fine with the original E5520s installed. Anyone know what the issue could be? Only possible thing I could think of would maybe be power supply issues, as the X5690 CPUs use an extra 100W combined, although I haven't read anything related online.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
X5690 Xeons not booting in Mac Pro 4,1
- Thread starter Marmotta
- Start date
- Sort by reaction score
Power is not an issue but you'll have some extra fan noise caused by 130W TDP per CPU.
I assume that CPUs still have their heat spreaders on if you have tried them in 5,1?
Dual 4,1 is designed for lidless processors, so you'll have to be extra careful when attaching your heat sinks back on. If you put them too loose Mac won't post, if you screw it to much you'll ruin CPU sockets. It's a tricky thing to do.
Also, you'll have to detach heatsink connection and put it by hand on processor board.
I assume that CPUs still have their heat spreaders on if you have tried them in 5,1?
Dual 4,1 is designed for lidless processors, so you'll have to be extra careful when attaching your heat sinks back on. If you put them too loose Mac won't post, if you screw it to much you'll ruin CPU sockets. It's a tricky thing to do.
Also, you'll have to detach heatsink connection and put it by hand on processor board.
Not really you can push it down into its slot by using smallish screw driver of it does not seat properly and the tighting is the key or was when I did same the mod few years back.
As owbp said, it's tricky, and you have to do it step by step.
Install CPU A first, then try to boot (fan will run at full speed), only when you can boot to desktop with all associated 4 RAM stick avail, then install CPU B.
Install CPU A first, then try to boot (fan will run at full speed), only when you can boot to desktop with all associated 4 RAM stick avail, then install CPU B.
Just be sure before proceeding that you didn't bent pins on CPU socket in previous attempts.
That would be a serious problem.
That would be a serious problem.
Oh, God, no. I've been very careful with the pins and, as I said, it runs fine with the old E5520 CPUs in it. If I'd done anything like that to the sockets, I'd probably be looking for a new CPU tray on eBay.
I have owned a 4,1 before and upgraded to 2x5650. as above, you must to know the pressure of screw to lock the CPU, even I have removed many times, I need to balance how is the best position to lock the Cooling Fan and CPU. Otherwise tones of error will happen (Ram alert, no boot screen... etc) and h9826790 is correct too, the MP model, you should install CPU A only first to test if it can boot. Then you install CPU B as well.
After my 4,1 backplane dead, I found a good condition 5,1 and use back my old stuff. Firstly it never boot when I put 2x5650 inside. Finally, I put one 5650 in CPU A first and it boot, then install CPU B and reset PRAM. Machine is return and perfect! for your 4,1 how you lock the Cooler screw is the key. More try!!
After my 4,1 backplane dead, I found a good condition 5,1 and use back my old stuff. Firstly it never boot when I put 2x5650 inside. Finally, I put one 5650 in CPU A first and it boot, then install CPU B and reset PRAM. Machine is return and perfect! for your 4,1 how you lock the Cooler screw is the key. More try!!
The same thing happened to me when I was upgrading my 4,1 with X5670s. CPU A by itself let the machine boot, but I didn't get any response (no chime OR video) after adding in CPU B.
In my case, the CPU B heatsink was too tight. I was using Xeons that I had de-lidded myself; but, even then - with CPUs of normal height - my tightening of the heatsink was overdone. I went and loosened each screw on the CPU B heat sink (all the way) and then re-tightened them - this time, 'finger tight' instead of completely tight. The machine booted fine after that and continues to function well.
I hope this helps! Pray that one of the CPUs / CPU sockets isn't defective. That's a possibility.
In my case, the CPU B heatsink was too tight. I was using Xeons that I had de-lidded myself; but, even then - with CPUs of normal height - my tightening of the heatsink was overdone. I went and loosened each screw on the CPU B heat sink (all the way) and then re-tightened them - this time, 'finger tight' instead of completely tight. The machine booted fine after that and continues to function well.
I hope this helps! Pray that one of the CPUs / CPU sockets isn't defective. That's a possibility.
