MP 1,1-5,1 5,1 runs like butter . . . that is when it starts . .

[edenorchestra]

Let me try to be concise:
Original box Mojave, cMP 5,1 2010, 6core-single 3.33, 24GB, Radeon7970, SSD boot, and assorted storage.
PCIe: RME RAyDat Audio, UAD-Audio DSP.

I have power up issues, you have to "play 3-4 minutes" with the push button to start (for years), after that it runs fine. Decided to upgrade, I'm pretty tech savvy, good test gear/tools etc, have a degree in EE, and love tech. This is in a recording studio.

Replaced the PS with a used eBay unit and a Radeon Vega 56 running 64 FW w/pixel mod. Added Sonnett USB-C 4 port, and dual 6 core x5960's, 96GB RAM. Seemed at first that it booted normal again, not much left of the original cMP. I also swapped the UAD for the Sonnett, works perfect. Had a utility power cycle, then the intermittent starting appeared to be back. At least in the shuffle it appeared that way, perhaps I got lucky and it started a few times normal. In any case it just started yet again. BTW everything's now on a 1500VA UPS, my bad.

Continuing on I replaced the backplane, and still have the same issue. Created a bootable Mojave NVME, IOCrest with (2) EVO 970+, all checked out, Boot ROM, NVME firmware etc. Runs like butter . . except it won't start consistently. Could I have gotten two bad supplies, or backplanes, yes could have. Thanks for listening . . this is a mystery, maybe the case? (Joking)

 

[edenorchestra]

BTW I staged the startup hardware, Video card, audio cards, SSD boot then fresh installed Mojave on a virgin EVO, nothing changed.

 

[casperes1996]

What behavior does it exhibit when it doesn’t work? Does it seem to get power but just not boot? Do you have fan spin? Could the power button itself simply be unreliable?

It seems like you’ve checked nearly everything really

 

[KeesMacPro]

Did you test it with minimum configuration e.g. 1 HDD/SSD (with a clean OS) in a SATA bay, all PCIe cards removed,all USB peripherals removed ,I/OCrest removed, 1 monitor connected, a wired keyboard and mouse?

Did you replace the front panel board with power button and usb ports?

In case you havent downloaded the Technician Guide for Mac Pro 2010:

http://tim.id.au/laptops/apple/macpro/macpro_mid2010.pdf

This is a fragment of the Technician Guide about the described issue (note point 4):

 

[Macschrauber]

There are start pads on the backplane where the self test switch is. You can (but not have to) plug off the front board cable and start via shorting the pads.

Also the Mac Pro should start immediately when pulling the clock battery

 

[KeesMacPro]

There are start pads on the backplane where the self test switch is. You can (but not have to) plug off the front board cable and start via shorting the pads.

Sounds like screenshot post#4...

 

[edenorchestra]

What behavior does it exhibit when it doesn’t work? Does it seem to get power but just not boot? Do you have fan spin? Could the power button itself simply be unreliable?

It seems like you’ve checked nearly everything really

Once its up it runs perfect, no issues. when you push the button you can hear the PS relay close and then it clicks a second time.. Almost as if its trying to power up and then drops back out. No fans at that point in time, just the clicks. Then after a handful of playing around it finally boots. Its been running all day now just perfect. I hate to have to shut it down.

 

[edenorchestra]

Did you test it with minimum configuration e.g. 1 HDD/SSD (with a clean OS) in a SATA bay, all PCIe cards removed,all USB peripherals removed ,I/OCrest removed, 1 monitor connected, a wired keyboard and mouse?

Did you replace the front panel board with power button and usb ports?

In case you havent downloaded the Technician Guide for Mac Pro 2010:

http://tim.id.au/laptops/apple/macpro/macpro_mid2010.pdf

This is a fragment of the Technician Guide about the described issue (note point 4):

Yes test with min config. Clean SATA, only video, actually my previous 7970 to see boot screens. just like you said, the bare minimum to get started. I have the tech guide I will look closer. Thanks for you input!

 

[edenorchestra]

There are start pads on the backplane where the self test switch is. You can (but not have to) plug off the front board cable and start via shorting the pads.

Also the Mac Pro should start immediately when pulling the clock battery

Ok cool some more things to try, Thank you.

 

[edenorchestra]

Sounds like screenshot post#4...

I'm on it . . will report back. Thanks everyone.

 

[ADDvanced]

I have an almost identical setup.

3.33 hex/32 ram/7970/ssds and Nvmes and hdds.

It runs great. Sometimes when I boot back into OS X from windows it just hangs. I’ve waited 5-10 minutes and it’s just a blank grey screen after selecting the OS X start up disk from the boot screen. Eventually I power down and select the same OS X disk and then it works fine.

This has happened twice. I’m running Mojave.

 

[Macschrauber]

Sounds like screenshot post#4...

Yes, sometimes old dogs bark to quick

 

[casperes1996]

Can I just say that this is one of those threads where the OP provides really solid information, describes their issues and steps taken and everything in a good way and responds well and nicely to comments - I have no good ideas not already posted, but I want to show appreciation for that after having seen my share of threads that are just

"Help I have issue??!"

So thank you, OP.

 

[Macschrauber]

I dont understand the play with power button thing.

Does it say: fiddle with it a while until the machines comes to life?

Or power on - forced power off - power on etc etc?

 

[edenorchestra]

Can I just say that this is one of those threads where the OP provides really solid information, describes their issues and steps taken and everything in a good way and responds well and nicely to comments - I have no good ideas not already posted, but I want to show appreciation for that after having seen my share of threads that are just

"Help I have issue??!"

So thank you, OP.

Thanks man, I appreciate that! My vocation is designing and starting up three phase power and real time automation system 1/2 my brain, the other is playing and composing media. Its just best to make one's point efficient, too easy to get wrapped up in wasted words. This is a good group, I've been reading posts for weeks, then implemented what you all have refined.

 

[edenorchestra]

I dont understand the play with power button thing.

Does it say: fiddle with it a while until the machines comes to life?

Or power on - forced power off - power on etc etc?

So yes fiddle with it a while until the machine comes to life, exactly!

No today I've really cleaned, polished and tweaked the entire system. It runs superb for sure, everything except HD audio, but I'm also working on my ill struck audio mixer. Later for that, it should be fine.

 

[edenorchestra]

FOLLOW UP 1

To clarify, "play with the button" means I fiddle with it until the Mac final turns on

Plugged in a USB powered LED on the front panel today. Watching the DIAG LEDS I have 5V standby LED on utility plug in.
Pushing the power button sequence: USB LED goes on, PS relay clicks, times out say 6-7 sec, USB LED off, then on my next press it started right up.

Watched the backplane DIAG LED's on a normal startup, all is in concert with the manual. I do notice the GPU OK LED never comes on, but that seems irrelevant on operation. Maybe because the Vega64 has no boot ROM, guessing.

So weirdly today was fine, with the USB LED plugged in. Someone on this forum made mention of trying that, voodoo perhaps?

 

[Macschrauber]

ok, so maybe the button.

There is a tiny cable going from the button to the front panel. CAREFULY pull it out and try the Pads on the Backplane to start.

Maybe the button has some resistance when it should be open.

If you have good soldering skills you maybe solder a switch button to the pads for investigation.

 

[edenorchestra]

ok, so maybe the button.

There is a tiny cable going from the button to the front panel. CAREFULY pull it out and try the Pads on the Backplane to start.

Maybe the button has some resistance when it should be open.

If you have good soldering skills you maybe solder a switch button to the pads for investigation.

Thanks! I can do that. It was sleeping since last evening, I came home this afternoon and instead of the power light slowly pulsing as in sleep mode, it was full on. The computer would not start via wired Mac keyboard. Tried a bunch of combinations to start it, long press, short press, unplug and press. Finally as usual it just started on the button press. Thing is when you push the button, anything with a light on it, ie. USB drives etc, light up, but no full power up, no fans, nothing. Then to turn off I must long hold the power button ~4-5 sec. and then the LED's go out, you can then repeat that sequence. Eventually it starts.

 

[edenorchestra]

FOLLOW UP 2
Bypassed the button and used the pads to start same symptoms.
Did a careful minimum hardware startup as per Apple's tech manual.
LED STBY lamp is on when you plug it in.

So I'm starting each time by killing 120 VAC power, (using a power strip switch) and a USB Drive w/LED on any USB port for a visual. This sets a baseline for observation, and a working computer should always be able to start this way.

1. Momentarily push the power button, PS relay clicks, only the USB-LED lights up, nothing else.

2. Seven seconds or so, the PS relay clicks off the USB-LED goes out.

3. Repeat a few times, maybe 4 or 5 times then . .

4. Momentarily push the power button yet again, PS relay clicks, only the USB-LED lights up and STAYS ON, the PS relay does not click off, no other devices power up.

5. Hold in the button this time, LED goes out after six seconds or so, I keep holding another 10 seconds and let go.

6. Momentarily push the button, the computer starts up and runs like butter.

Damn . .

 

[eicca]

It sounds like a communication issue between the logic boards and the PSU. The CPU tray is the only thing you haven't replaced in that chain, so I'd give that a try.

 

[edenorchestra]

It sounds like a communication issue between the logic boards and the PSU. The CPU tray is the only thing you haven't replaced in that chain, so I'd give that a try.

Done that a while ago, there's nothing but the front panel and the case, I must have gotten a defective power supply, it behaves the same as the original. AHT/ASD all check out.

 

[Macschrauber]

Have you plugged off the front board during the tests with the pon pads?

 

[edenorchestra]

Have you plugged off the front board during the tests with the pon pads?

I don't know about this, pon pads? I'll search this, thank you.

 

[edenorchestra]

OK yes I have done that, didn't recognize the pon pad term. I have disconnected the push button and shorted the pads, the problem is the same, intermittent starts, runs great once its does. Maybe I should flip the case on its side to let the electrons flow easier