Low voltage from DisplayPort/VGA adapter from MacPro 2010 (on multiple video cards)

Question

Level 1

5 points

Low voltage from DisplayPort/VGA adapter from MacPro 2010 (on multiple video cards)

OK, here is a puzzle for hardware types.

My laboratory uses a MacPro 2010 to drive a VGA CRT monitor (via DisplayPort VGA adapter) and a DVI monitor (through a DVI connection). A few weeks ago, we had an incident where the power cable to the CRT monitor was severed by some equipment, sparks flew, and, subsequently, the VGA CRT would only work intermittently through the DisplayPort adapter (sometimes great, sometimes not functional).

The behavior was quite interesting and has me puzzled, and maybe someone can explain it.

1) The CRT itself is fine; other computers drive it well, so I know of no problem with the CRT.

2) When I monitored the voltage of the vertical refresh signal on an oscilloscope, I noticed that the pulses would sometimes be as large as 4V, and sometimes as small as 1V. When the pulse voltage gets low, the monitor doesn't work. It will oscillate in and out of this working/non-working mode (sometimes work for a few hours, sometimes not work for several hours). Other computers give a very reliable 5V signal.

3) I assumed the problem was the video card. So, I exchanged video cards with another Mac (I exchanged an ATI Radeon 5870 and an ATI Radeon 5770). Oddly, the other Mac (a MacPro 2012) could drive the CRT very well with either video card (its native one or the one that was exposed to the surge), and the MacPro 2010 had the same low voltage problem with the DisplayPort adapters on either video card.

4) Finally, I tried the DVI port on the ATI Radeon 5770, using a VGA to DVI adapter, and it works fine. Great 5V signal on the vertical refresh, rock stable image. The DisplayPorts on the very same card will not work.

5) A check of the major voltages on board with Hardware Monitor, and all looks well.

(Parameter RAM was reset, computer unplugged for hours, etc.)

So, what could cause DisplayPort output to be low voltage in a way that is independent of the video card but dependent on the computer? All other functions of the computer seem perfect (DVI monitors are fine, fine even through the DisplayPort). Is there some regulator on the mother board or power supply that has privileged access to the DisplayPorts on the video card but avoids the DVI port? Seems crazy, no?

Best

Steve

Posted on Oct 9, 2015 1:20 PM

Reply

Answer 1

Oct 9, 2015 2:39 PM in response to steve_vh

The signal you are monitoring is indeed intended to be a TTL-level signal. I think it may be negative-true. Data signals are typically 0.7V peak-to-peak.

You must be very cautious in using AMD/ATI graphics cards when multiple monitors are attached, and it is thought that the power falls off with multiple displays attached, and so quickly that not more than TWO displays can be supported at once without ACTIVE (read powered) adapters.

Reply

Answer 2

steve_vh Author

Level 1

5 points

Oct 9, 2015 5:11 PM in response to Grant Bennet-Alder

Thanks for your input. In this case, the issue is computer specific (not even video card specific). I only have 2 monitors attached, 1 VGA and 1 DVI, through separate adapters. When I use the MiniDisplayPort to VGA adapter on the computer that got the surge, I don't get full voltage. If I take that same video card and put it in a different computer, I get full voltage. The DVI port (not the MiniDisplayPort) seems fine. I'm really puzzled by it.

Thanks

Steve

Reply

Answer 3

Oct 9, 2015 6:23 PM in response to steve_vh

When all else fails...

...make a chart of your findings.

You may see that you have discovered the truth, but hidden it in some way (usually by making assumptions.)

Sometimes I find that there is a case I forgot to test, and its results hold the key to understanding.

Reply