Dell PowerEdge C8220 Build and Questions

drabadue

Active Member
Sep 15, 2016
162
54
28
@Wictar - for some reason, that jumper isn't working for me. I tried both SATA ports and can't get either one to recognize. It keeps telling me "No bootable device". Did you chance settings in your BIOS to get it to boot windows?
 

Wictar

New Member
Sep 19, 2016
27
13
3
34
@Wictar - for some reason, that jumper isn't working for me. I tried both SATA ports and can't get either one to recognize. It keeps telling me "No bootable device". Did you chance settings in your BIOS to get it to boot windows?
Yes, I reconfigured BIOS. In "Boot" tab turn "Force PXE Boot only" to "disable", change in "Boot Type Order" "Hard Disk" on the top. To determine the contact you want to close, I charted voltage pins. Then checked different combinations, by closing "+" and "ground". Multicore computer do I need for the job. For the rendering of architectural exteriors.
 
Last edited:
  • Like
Reactions: ComradeMIKE

drabadue

Active Member
Sep 15, 2016
162
54
28
Yes, I reconfigured BIOS. In "Boot" tab turn "Force PXE Boot only" to "disable", change in "Boot Type Order" "Hard Disk" on the top. To determine the contact you want to close, I charted voltage pins. Then checked different combinations, by closing "+" and "ground". Multicore computer do I need for the job. For the rendering of architectural exteriors.
I believe I have the same settings in BIOS, but that jumper is not working for me. I must have something else wrong. Do you know what version board you have? I believe mine is v1.0 and i believe the other versions are v1.1 and v1.2. The later version boards have more connectors on them. I would not think the front panel connector would change between the versions though.

Do you have a chart of the front panel voltages and the changes when grounded?

I will try and experiment with mine more tonight.
 

Wictar

New Member
Sep 19, 2016
27
13
3
34
Motherboard Name is Dell DCS7200S (information from AIDA64). You can to try your own jumper combination, I think. I tried different combinations at their own risk, nothing burned:) See my PDF below, this is voltage scheme.
 

Attachments

Last edited:

drabadue

Active Member
Sep 15, 2016
162
54
28
Some updates that I have discovered playing around a little more with the system.

I guess you will have to drill custom holes into the case to mount the MB?

Also what's the specs on those pcie slots? just want to make sure that they have adequate wattage supply for gfx cards. are they at least 8x 2.0? will have to get riser card or ribbon cable for them...

It's workable but there's def a fair amount of custom work that will need to be done.
@wildpig1234 - I have discovered the PCIe ports are v2.0. See the image below:



(For some reason the image wont show up embedded, so here is a link to it:)
https://dl.dropboxusercontent.com/s/iionsjluu99iq6d/IMG_20160922_005336.jpg?dl=0

Regarding this image, I wanted to bring up a few more interesting things.

My motherboard model number is 0W6W6G.

USB Port Count: 14 ports

I know the board has 2 USB ports in the front. I had a mini WiFi adapter and my wireless keyboard & mouse combo plugged into those ports. I also know there is a 10-1 motherboard header that is for USB. Now, that would give me 4 ports total. I am wondering why there are 14 ports listed? In another software (I failed to get a picture of it), there were two different USB Root Hubs listed. The one hub had ports 0-6 and my WiFi adapter and keyboard/mouse were listed in two of those ports. The other hub had ports 1-8 and there were nothing listed as being attached. I assume the second hub is the internal header.

----------------

@Wictar - Thank you for making a clearer pin out of your drawing. I have made my own pin out for the wiring colors that come off of the front panel connector harness. I am not sure the wire colors are significant, but figured it wouldn't hurt to be documented. My drawing is upside down compared to yours. Take a look at the position of the alignment notch:



The purple box shows the wires that we are jumping to get the board to power on. I have confirmed that the top left (black wire) is ground on my board. I have not checked the other pins yet.

Have you tested the pins for voltages at both the on and off states? As in, test each pin with the board on and then test each pin with the board off but power still applied?

----------------

Here is a picture of the routing of the front panel harness:



All 12 wires are populated in the harness and they all run to the Power Distribution Board.

----------------

I also found out the board does not have any type of on board audio chip. I believe the graphics card I will be using (GTX 780) should have an on board audio chip and should be able to output audio over HDMI or Display Port.

I have also ordered the PCIe Ribbon cable for the GPGPU that extends off the back of the board. I will hopefully have them sometime next week and see if I can get the graphics card working as well as audio.

----------------

Back to this RAM:




As it turns out, the 4 sticks are 16GB total. Apparently since the tag says 8G and "KIT OF 2" That means they are 4GB modules.

With that said, I bought some more RAM off eBay. They are still Kingston KVR, but they are 1333MHZ instead of 1066MHz and they have different timings. I hope they will play well together.

----------------

I have reached out to a few computer savvy friends of mine to try and help figure out the rest of the front panel connectors. Hopefully we can find a PS_ON pin so the power supply doesn't have to run at all times.
 

drabadue

Active Member
Sep 15, 2016
162
54
28
I have yet to confirm that these fit, but I believe these may be some useful part numbers for the connectors if we wanted to make custom wiring harnesses.

12 Pin Front Panel Wire Side Connector
Molex Milli-Grid Housing: 51110-1251
Molex Terminals: 50394

18 Pin Motherboard Power Wire Side Connector
Molex Micro Fit 3.0 Housing: 43025-1800
Molex Terminals: 43030

2 Pin HDD Interposter Board Power Wire Side Connector

Molex Micro Fit 3.0 Housing: 43025-0200
Molex Terminals: 43030

4 Pin Disc Drive Power Connector (male terminal side to tap into the +12v and +5v power supply)

AMP Housing: 1-480426-0
AMP Terminals: 60620-1
(It seems Molex no longer makes the male terminal side)


For the mezzanine cards and rear interposer/PCIe slot, I believe these are one of the following:
Molex EdgeLine CoPlainer
Molex EdgeCard Mezzanine
Molex SpeedEdge

I want to find out what style terminal they are to see if I can make or buy a riser cable to then relocate these ports to a usable location. I am going to try and get in touch with someone who can help identify what connectors these are.
 

drabadue

Active Member
Sep 15, 2016
162
54
28
I think I might have made a little more progress with the front panel connector. I did a little more testing and found some interesting characteristics on a few pins.



Once again, we are looking at this drawing with the alignment notch on top, and the ground in the top left corner.

I played with the board in 3 different states. The first state was 12v stand by line receiving power, all grounds connected, but no other 12v power on the power connector.

This would be the "Standby" state.

When I did this, I had a power supply that was outputting 13.8v to the standby line. It was all I had, so I made due. In this state, I had an amber LED blinking on the power button.I found out later in the BIOS logs that the standby voltage was too high and set a trigger to blink the power button LED amber. When checking the voltages for the front panel connector, I noticed some interesting voltages changing on the pin that I have listed as "BMC?" above. The voltages would go from around .11v to 3.34v and it would bounce back and forth. Maybe it was for the blinking LED? I need to do more research on this pin.

I also noticed that the top right pin (circled in blue in the drawing above) was at 0.01v, and the pin circled in yellow was at 3.34v. These will come into play later.

The next thing I tried was powering all the 12V pins in the power connector, but leaving the board off (no jumper and not pressing the power button).

This would be the "Power on Board off" state.


I did this with the ATX power supply hooked up, but left the jumper out. When checking the voltages, I noticed something interesting. The top right connector circled in blue, was now 4.97v. And the pin circled in yellow was still 3.34v.

Next, I jumped the pins on the front panel connector (the ones circled in purple above).

This would be the "Power on Board on" state.

0v/5v

As soon as the jumper was inserted, the 5V pin (circled in blue) instantly dropped down to 0v. When I pulled the jumper out to shut down the board, the pin stayed at 0v. This is with all 12v pins on the power connector still receiving power. I then flipped the switch on the power supply to kill power to the power connector. Then I turned the power supply back on via the switch and the voltage instantly went back to 5v on this pin. Plug the jumper back in, the pin goes right back to 0v.

I then decided to check the resistance on this pin to ground. The pin is normally open with the board off. Plug in the jumper or push the front power button with the jumper plugged in, the pin goes to about 13 ohms. And the pin will stay at 13 ohms after that until the power supply is physically turned off via the switch on the back at which point it returns to open.

This leads me to believe that this pin might be our PS_ON line. I think we can try hooking this pin to the green wire on the ATX power supply. I am going to order a 12v Booster to up the voltage on the ATX 5v standby line to 12v so that the board has the right amount of power when the power supply is off. That way I can try to hook this front panel pin to the PS_ON line of the ATX power supply to see if I can get the power supply to turn on and off with the power button automatically.

I also tested this pin when in Windows 10. With Windows 10 running, the pin is at the 13 ohms. When you click shut down, Windows will shut down and shortly after the pin will go back to open. Just another reason I think this might be the PS_ON pin.

0v/3.3v
The other interesting thing I noticed in this state is the pin circled in yellow. When the power supply is turned on and the jumper is out, the pin has 3.34v on it. Wait about 15 seconds and the voltage drops to 0.11v and stays there. Wait another 30 seconds and the voltage goes back up to 3.34v. It may be a coincidence, but about 5 seconds after that and the BMC LED begins to blink.

As soon as the jumper goes in, the pin goes to 0.11v. Take the jumper out and the pin goes back to 3.34v. I need to do some more looking into this, but I found it to be rather interesting.


I still have yet to figure out how to get the 2 on board SATA ports working. I have tried the jumper @Wictar suggested. I will continue to try and see if i can get them to work. I am also going to order the connectors, 12v booster, and pins that I need to try the standby/PS_ON idea.
 

Wictar

New Member
Sep 19, 2016
27
13
3
34
I have a lot of work. I'm already working at a new computer:) . I`ll try to help you to fix the problems later.
 

drabadue

Active Member
Sep 15, 2016
162
54
28
No worries. I am going to keep working on it to see what I can do. What's your new computer? Any reason for starting on another one?
 

Wictar

New Member
Sep 19, 2016
27
13
3
34
No worries. I am going to keep working on it to see what I can do. What's your new computer? Any reason for starting on another one?
The new computer is very productive for its price. For rendering in 3DS Max - Great.
 

sdip

New Member
Sep 27, 2016
1
0
1
28
This is great info! Currently these boards are the cheapest dual LGA2011 on ebay. Is the front panel connector yet another fancy Dell thing?
 

drabadue

Active Member
Sep 15, 2016
162
54
28
This is great info! Currently these boards are the cheapest dual LGA2011 on ebay. Is the front panel connector yet another fancy Dell thing?
I have confirmed the PS_ON front panel pin. It is in fact the pin circled in blue in the drawing above. I have boosted the +5vSB line from the ATX power supply to +12v using an LM2577 Boost Converter. Once I had the standby line up to 12v, I tested my theory on the PS_ON pin mentioned in previous post. It works hooked up to the GREEN wire on the 24 pin ATX connector. The board will turn on and off via the power switch on the board and windows will even shut down the board.

I have also confirmed the part numbers for the connectors listed above. The mini fit and micro fit connectors are correct.

Also, the mezzanine connectors are 3M SPD08 High-Speed Card-Edge connectors. They are not molex. The rear connectors for the SATA interposer and the rear PCIe x16 are 120 pin connectors where the front mezzanine I believe are 100 pin.
 

drabadue

Active Member
Sep 15, 2016
162
54
28
Something else worth mentioning. I read in a Dell forum that the C8220 motherboard V1.0 and V1.1 will not work with the E5-2600 V2 processors. I am not sure why, but they said the C8220 V1.2 boards would work fine with V2 processors.

The easiest way I know how to tell the difference is looking at the SATA ports. The V1.0 board has 2 SATA ports by the mini SAS connector. The V1.1 and V1.2 boards have 4 SATA connectors. Two by the mini SAS and two on the rear of the board near the front panel connector.
 
Last edited:

thomasz

Member
Oct 2, 2016
46
43
18
40
@drabadue and @Wictar, you guys are doing an awesome job! I followed your guidelines and successfully powered up my C8220! Thank you so much! Here is my picture:

Meanwhile, my board (v1.0) is capable of booting from a regular SATA hard drive thru its onboard SATA port. I didn't change anything in BIOS or apply additional jumpers.

By they way, do you all like my coolers? :)

01.jpg
 

MiniKnight

Well-Known Member
Mar 30, 2012
3,000
910
113
NYC
The easiest way I know how to tell the difference is looking at the SATA ports. The V1.0 board has 2 SATA ports by the mini SAS connector. The V1.1 and V1.2 boards have 4 SATA connectors. Two by the mini SAS and two on the rear of the board near the front panel connector.
So how do you tell V1.1 from V1.2? That seems to be the big question for me.
 

drabadue

Active Member
Sep 15, 2016
162
54
28
@drabadue and @Wictar, you guys are doing an awesome job! I followed your guidelines and successfully powered up my C8220! Thank you so much! Here is my picture:

Meanwhile, my board (v1.0) is capable of booting from a regular SATA hard drive thru its onboard SATA port. I didn't change anything in BIOS or apply additional jumpers.

By they way, do you all like my coolers? :)

View attachment 3521

@thomasz Looks good. Glad to see another one fired up. It looks like your board is similar to the board @Wictar has. It looks like neither of you have the card edge slots at the rear of the board for a hard drive interposer. Interesting that you got yours working with no jumpers. Can you post pictures of the BIOS settings for the SATA ports? Perhaps I'm missing something.

So how do you tell V1.1 from V1.2? That seems to be the big question for me.
@MiniKnight I believe you can tell the differences visually. I'll study the manual and get back to you.
 

drabadue

Active Member
Sep 15, 2016
162
54
28
I've actually played with mine a bit more. Currebtly trying to use the rear card edge PCIe slot for a graphics card, but coming up empty handed. The card shows up in device manager with the correct model number. I have the correct drivers installed. It's coming up with error code 43 in device manager. See the pictures below.

Here are the gpgpu extender cables:


And they have a power connector on the back side of them for +12v and ground. I have that wired up to a 4 pin sata power connector from the power supply.

Here is the setup with the graphics card:


And here is a picture of the error code in device manager:


I have tested the PCI slot with a wifi adapter and it connected me to the internet. So I know the slot at least works.

Now, the strange thing is, I took the same hard drive I was using here, and the same graphics card, and plugged then into my desktop PC motherboard and the graphics card was recognized and worked perfectly.

I wonder if I am missing a motherboard driver or a setting in the BIOS in order to get the graphics card working. If anyone has any suggestions, I'm all ears.
 

Spun Ducky

New Member
Sep 20, 2016
15
0
1
36
The error 43 is usually related to drivers, defective card, and or insufficient power. I would try a few different drivers first then from their confirm the 12v rail you are tapping into the card has sufficient amperage. This was a message that I have ran into with egpu solutions on laptops.
 

drabadue

Active Member
Sep 15, 2016
162
54
28
The error 43 is usually related to drivers, defective card, and or insufficient power. I would try a few different drivers first then from their confirm the 12v rail you are tapping into the card has sufficient amperage. This was a message that I have ran into with egpu solutions on laptops.
@Spun Ducky - Thank you for the advice. I downloaded the latest drivers from Nvidia and I tried installing and reinstalling them to make sure they were right. The strange thing about installing those drivers, is that the Nvidia Control Panel never came up when you right click on the desktop. Normally you can right click anywhere on the desktop and there will be an Nvidia Control Panel shortcut to allow you to adjust your resolution and color settings. That never showed up on this motherboard. I even tried to uninstall that driver and then install an older driver, just to see if it made a difference - same result.

The card is powered by an eight pin and a six pin PCIe connector from the power supply. I know those are good because there is an LED on the video card near those power connectors that will turn from red to green once they are plugged in. Also, the other system that I installed the SSD and GPU into worked with the same power supply and the same driver. I literally changed nothing and they just flat out worked.

The only power variable that I might have is the 4 pin power connector on the PCIe riser itself. There is a 4 pin power connector that is labeled as 12VPWR. I tested with my ohm meter that 2 of those terminals were ground, and I *assumed* the other 2 were +12v. Watching YouTube videos of people taking these boards apart showed the wires connecting to this connector as 2 yellow and 2 black. Which lead me to believe it was 2 +12v and 2 ground. Perhaps something in that assumption was wrong. I really wonder if I even need to power the PCIe riser since the card gets its 12v power and ground directly from the PSU?

If all goes well, hopefully tonight into tomorrow I can test this setup again with a fresh install of windows just to start from scratch and eliminate all other variables.