F4-220 - Start problem

[MaksDampf]

I googled some more and some LPC headers also have a Data pin on Pin1. So it is advisable to further test it before.

It might be of help, that the Clock pin normally should read out around 1,5V with a multimeter.
So the CLK pin should read something between 1 and 2Volt. 1.5V is normal, but its probably lower which is why it is faulty.

Sadly the Data pins should read out 3.3v as well, so it is easy to confuse them with 3.3v Power. they just can't supply any current and voltage will probably drop once you connect a resistor to it (and maybe damaged). Maybe get 3.3v from somewhere else then (not the LPC connector)?

[kavdiev]

If you take my last picture. On your left is pin 9 and right is pin 1

When power off.
Pin 6-l and 7 no voltage
All others 8.2mV

After power on
1 - 3.359v
2 - 3.355 -> 357
3 - 3.354 -> 356
4 - 3.355 ->355
5 - 3.355 -> 356
6 - 0.2 mV so let's say 0
7 - 3.359v
8 - 1.892v
9 - 3.359v

Now I need to fin a resistor

[kavdiev]

Ok.
Pin 6 is ground

[MaksDampf]

OK, so the clock pin is very likely Pin8. As for 3.3V this is difficult to figure out.
- there should be 4x data pins, which could be any of the 3.3V pins.
- Additionally there are 3V standby. you should be able to measure 3V there when the device is off but on PSU.
- then there is a Reset and Lframe pin which should also read 3.3v.

I don't know if it will damage the board if you put the resistor to one of the data pins or reset by accident.

[MaksDampf]

I had a look at your pictures again. Pin6 is very clear GND also from the backside of the PCB it is visibly connected to the groundplane on the back.

On the Pin9 there is a large capacitor, which is probably your 3,3V power.

on Pin8 there is a capacitor too, which is strange for a clock signal. Also 1,8V would be a good voltage level for the clock.

in qnap forums they say if its above 2.4V you should pull it to gnd with 100ohms. if its below 1V you need to pull it up, which is more often the case with synology and apparrently also supermicro/cisco hardware.
https://forum.qnap.com/viewtopic.php?t=157459

[MaksDampf]

About the chance of burning something when connecting is to the wrong pin: 100Ohm is 100mw or 30Milliamps on 3,3V. This is safe for most µCs like an Arduino for a few seconds. raspberry PI i/o is limited to 20ma per Pin, but it should be safe for a few seconds.

But the Intel Atom is a chip at a much different process node has is probably several times lower current support per pin. So if you try it out by testing the pins, try to keep it very short or start with a 1K or 470R then 220R, then 100R Resistor! The current will be 10 times lower with 1K, but also the pullup/down will be less effective.

[kavdiev]

pin 8 did not helped

tried with others no result


by the way here is a picture of Frontpanel of F4-220 (version 2018) so it is actually F5 motherboard inside F4 box.
i think i've already mentioned that before , here you can see 5 diods for hdd + 1 LAN + lan2?? +1 power
an here is a picture of CPU without radiator

[MaksDampf]

ty for the pictures.

did you try pulling it up to 3,3v or down to gnd or both?

[kavdiev]

Hello,
i need some explanation what is pulling down or up ?
i connected 1k and after some others and 100 Ohm resistors to ping 8 and 6

[MaksDampf]

Well, depending on the actual hardware, the clock signal is either too high or too low. So for synology this means a resistor between clock and 3,3V - for mist qnap devices it seems to be the other way around and the clock signal has to be pulled down to gnd. (resistor between pin8 and gnd).