First SFF build. Cannot power on. Please assist.

Steedsofwar

Chassis Packer
Original poster
Oct 14, 2018
15
0
My system only has a 2400g, 2x 4gb Ram, and two 40mm noctua case fans. No LED strips or GPU. One 2.5" 256gb SSD. That's it. The wattage this system should sip is max 120w. AsRock b450 fatality mobo. Just refuses to power on.

I'm using Pico 160xt. But my power brick is 156w and doesn't match the 160w of the Pico. Will this be the issue?

I have managed to boot into BIOS just fine by ripping out the power cables from my main machine and transplanting the mini itx board manually. No issues at all. So I'm guessing it's my 156w power brick by Leicke or my Pico 160xt. I just dunno which one. Or if these two components are incompatible.

Please help.
 

Thehack

Spatial Philosopher
Creator
Mar 6, 2016
2,800
3,650
J-hackcompany.com
You should take a picture of your set up. Use imgur to host it.

If you have a multimeter, start by checking the output of your AC-DC adapter, your 156w brick.
 

Steedsofwar

Chassis Packer
Original poster
Oct 14, 2018
15
0
You should take a picture of your set up. Use imgur to host it.

If you have a multimeter, start by checking the output of your AC-DC adapter, your 156w brick.
Thank you for the reply. If you are going to be online, I can try and open the case and upload an image for your feedback. Unfortunately, I don't have a multimeter as I only just found out about it in one of the writeups here AFTER encountering my problem.
 

Steedsofwar

Chassis Packer
Original poster
Oct 14, 2018
15
0
Does it have to be a 12v adapter brick or can it be a 19v variant? In order to marry it to the internal Pico PSU?
 

Thehack

Spatial Philosopher
Creator
Mar 6, 2016
2,800
3,650
J-hackcompany.com
Okay. Seems like have the correct set up. Here are the trouble shooting steps.

1. Check output of 12V ac adapter. You will need to borrow a multimeter, or use a device that uses the same voltage and plug. Reading should be 11.5-12.5v.

2. Check solid connection to picopsu power, while being careful not to cross the probes (very important) check the solder points of the black and white wires. Use a continuity check between the solder point and the barrel plug. This ensures you have a solid connection.

3. Check function of power switch. Use a screw driver to short the power switch pins. Switches can break too.

4. Check the function of pico psu. Using a 24 pin atx chart, short the power_on and ground pin. Check the output of any of the 12V pin (in reference to ground) for output.

5. If all of the above has the correct reading and output, then it is a pc component issue.
 

Steedsofwar

Chassis Packer
Original poster
Oct 14, 2018
15
0
Thank you for your detailed steps. I won't pretend I understood it all but I am certainly gonna give it a try.

I've ordered a very budget multimeter to arrive tomorrow.

I believe I can carry out 1, 2 and 3 just fine. I may have to come back to you for No. 4 though.

I'm not familiar with continuity tests but I will look these up online.

Thanks again. I'll report my findings once I get the meter in.
 

Steedsofwar

Chassis Packer
Original poster
Oct 14, 2018
15
0
Use a continuity check between the solder point and the barrel plug. This ensures you have a solid connection.

Do I test the white solder point? And by the barrel do you mean the circle port that the adapter will insert to at the back of the case? And what reading should I be looking for? Should the multimeter be set for DC?

I have checked the adapter and it is showing 12v just fine. So I can eliminate that as the problem.
 

Thehack

Spatial Philosopher
Creator
Mar 6, 2016
2,800
3,650
J-hackcompany.com
Do I test the white solder point? And by the barrel do you mean the circle port that the adapter will insert to at the back of the case? And what reading should I be looking for? Should the multimeter be set for DC?

I have checked the adapter and it is showing 12v just fine. So I can eliminate that as the problem.

On the 150 xt pcb. You do a continuity check by setting it to Ohm or resistance. You will check for 0 ohms between the solder point on the pcb, and the power entry where you would plug in the AC adapter plug. If it reads OL, try the other solder point. There will be two solder points one for the black wiring and one for white wire.

The both the solder points should have continuity (0 ohms) between either the inside pin or the inner barrel of the power entry.

If you can't quite figure that out, check the DC voltage between the black and white solder point while the kit is powered on, with the AC Adapter plugged in. But this needs to be done carefully without touching the two test leads together. If you touch them you will short and cause damage somewhere.
 

Steedsofwar

Chassis Packer
Original poster
Oct 14, 2018
15
0
On the 150 xt pcb. You do a continuity check by setting it to Ohm or resistance. You will check for 0 ohms between the solder point on the pcb, and the power entry where you would plug in the AC adapter plug. If it reads OL, try the other solder point. There will be two solder points one for the black wiring and one for white wire.

The both the solder points should have continuity (0 ohms) between either the inside pin or the inner barrel of the power entry.

If you can't quite figure that out, check the DC voltage between the black and white solder point while the kit is powered on, with the AC Adapter plugged in. But this needs to be done carefully without touching the two test leads together. If you touch them you will short and cause damage somewhere.
Alright I will reply with an update soon as I can test it. I will try the continuity test. I understand that you want me to try both the solder points, one at a time of course and look for 0 ohms.
 

Thehack

Spatial Philosopher
Creator
Mar 6, 2016
2,800
3,650
J-hackcompany.com
Alright I will reply with an update soon as I can test it. I will try the continuity test. I understand that you want me to try both the solder points, one at a time of course and look for 0 ohms.

After you try that, if there is good continuity that means power should be reaching the pcb.

Next unplug 4 pin from the barrel power entry. Check the DC voltage, with the red test lead on where the white wiring plugs into, and the black lead where the black wiring plugs into. Do this while the adapter is in and plugged in. You should read +12V. This ensures that you are getting the correct polarity.

Finally, if that passes, use a screwdriver or other conductor to turn on the pc. This ensures that the switch is not the problem. If there is still no power on (led, fan spin, etc) and you have confirmed the motherboard boots with a regular atx psu, then the issue is the 150XT itself.
 

Steedsofwar

Chassis Packer
Original poster
Oct 14, 2018
15
0
After you try that, if there is good continuity that means power should be reaching the pcb.

Next unplug 4 pin from the barrel power entry. Check the DC voltage, with the red test lead on where the white wiring plugs into, and the black lead where the black wiring plugs into. Do this while the adapter is in and plugged in. You should read +12V. This ensures that you are getting the correct polarity.

Finally, if that passes, use a screwdriver or other conductor to turn on the pc. This ensures that the switch is not the problem. If there is still no power on (led, fan spin, etc) and you have confirmed the motherboard boots with a regular atx psu, then the issue is the 150XT itself.
Thank you again.

I can already confirm that the screwdriver method of switching on for power works as I have tried this when testing the full ATX PSU. I'll update you ASAP.
 

Steedsofwar

Chassis Packer
Original poster
Oct 14, 2018
15
0
Okay things have gone from bad to worse!

So I purchased an £8 Pico psu and a multimeter for not much more. The Pico PSU showed 12v when testing the two solder points using the MM, kit fully plugged in. I then checked them for resistance and it showed zero from one point and no reading from the other solder point. I shrugged and tested the adapter and it showed me 12v too. I was thinking...if everything is showing me the correct readings, except for one solder point, then this £8 Pico is doing just fine. The £40 Pico on the other hand was not showing me 12v when connected up to the adapter. The mm didn't beep for resistance testing either, between the solder point and the port that the adapter would plug into. The 4 pin plug carried the 12v just fine when disconnected from the Pico. It read 12v too.

I presumed that this pricey pico to be one dead and got ready to return it. It still didn't fix my problem though. I mean, the £8 Pico actually flashed the power led a single time. Every time I'd plug it in and every time I'd press the power button. But no fans spin.

Scratching my head, I decided to test it on my main psu like I did the last time. There is a setting in bios for this mobo that goes by the name of 'cool and quiet' or something along that line, when I was browsing through it in my last test. On further reading, the manual says this setting can affect SOME psus etc if left on the default enabled status.

I was certain that this was the culprit so I rerigged the 24pin and 4pin full ATX PSU plugs into the motherboard and pressed power to boot bios like last time. Only this time the CPU fan nearly spun but failed to move after like a slight jerk...For want of a better expression. I could smell something smoky and I quickly unwired everything. I rewired it all and this time the fan was spinning but no bios. I turned it off to try to power it again but this time, I was getting no fans spinning or anything at all. I could still smell the smoky aroma and packed everything in and gave up.

I think the mobo could have been the culprit. I really dunno. I don't wanna try and wake it again. I'm just too anxious that the mobo is quite literally fried. And could potentially fry my CPU and ram, if it isn't already dead too.

I'm still perturbed that the £8 PSU passed all the continuity and current tests bar one small point. And yet never powered up. I'm also even more befuddled as to why the ATX PSU would no longer power it up into BIOS and why it smelt like it is fried.

Looks like I need a new mobo? New PSU? Maybe it killed my ram and CPU, I honestly feel lost right now. I've built a couple PCs before and not had trouble at all, even though it was only just the couple times. Dunno what's gone wrong here this time. Feeling super despondent right now. Gonna sleep on it now if I can.
 

Thehack

Spatial Philosopher
Creator
Mar 6, 2016
2,800
3,650
J-hackcompany.com
Okay things have gone from bad to worse!

So I purchased an £8 Pico psu and a multimeter for not much more. The Pico PSU showed 12v when testing the two solder points using the MM, kit fully plugged in. I then checked them for resistance and it showed zero from one point and no reading from the other solder point. I shrugged and tested the adapter and it showed me 12v too. I was thinking...if everything is showing me the correct readings, except for one solder point, then this £8 Pico is doing just fine. The £40 Pico on the other hand was not showing me 12v when connected up to the adapter. The mm didn't beep for resistance testing either, between the solder point and the port that the adapter would plug into. The 4 pin plug carried the 12v just fine when disconnected from the Pico. It read 12v too.

I presumed that this pricey pico to be one dead and got ready to return it. It still didn't fix my problem though. I mean, the £8 Pico actually flashed the power led a single time. Every time I'd plug it in and every time I'd press the power button. But no fans spin.

Scratching my head, I decided to test it on my main psu like I did the last time. There is a setting in bios for this mobo that goes by the name of 'cool and quiet' or something along that line, when I was browsing through it in my last test. On further reading, the manual says this setting can affect SOME psus etc if left on the default enabled status.

I was certain that this was the culprit so I rerigged the 24pin and 4pin full ATX PSU plugs into the motherboard and pressed power to boot bios like last time. Only this time the CPU fan nearly spun but failed to move after like a slight jerk...For want of a better expression. I could smell something smoky and I quickly unwired everything. I rewired it all and this time the fan was spinning but no bios. I turned it off to try to power it again but this time, I was getting no fans spinning or anything at all. I could still smell the smoky aroma and packed everything in and gave up.

I think the mobo could have been the culprit. I really dunno. I don't wanna try and wake it again. I'm just too anxious that the mobo is quite literally fried. And could potentially fry my CPU and ram, if it isn't already dead too.

I'm still perturbed that the £8 PSU passed all the continuity and current tests bar one small point. And yet never powered up. I'm also even more befuddled as to why the ATX PSU would no longer power it up into BIOS and why it smelt like it is fried.

Looks like I need a new mobo? New PSU? Maybe it killed my ram and CPU, I honestly feel lost right now. I've built a couple PCs before and not had trouble at all, even though it was only just the couple times. Dunno what's gone wrong here this time. Feeling super despondent right now. Gonna sleep on it now if I can.

Testing continuity and voltage is test the basics first. In troubleshooting we're testing if there is power in and it is setup properly.

I hope you did not damage anything. If it was setup properly, you were able to use bios with atx psu, but not minibox, then likely is the mini box issue after we confirm it is set up properly.
 
  • Like
Reactions: Steedsofwar

Steedsofwar

Chassis Packer
Original poster
Oct 14, 2018
15
0
Testing continuity and voltage is test the basics first. In troubleshooting we're testing if there is power in and it is setup properly.

I hope you did not damage anything. If it was setup properly, you were able to use bios with atx psu, but not minibox, then likely is the mini box issue after we confirm it is set up properly.

UPDATE

As I've mentioned, I tested for continuity on the throwaway Pico and all was well. I tested the power brick and it too was reading the rated 12v fine. The £40 Pico plug wasn't showing the correct readings.

1. Using the expensive branded Pico plug, the system refused to power on.

2. I used the full ATX power supply to power it on and that went fine. BIOS was accessible. So I ordered a cheap £8 Pico as a test replacement for the £40 Pico.

3. Cheap Pico also fails to power on the mobo. So now I'm super confused. Did I receive two faulty Pico psus?

4. So i try again to power up the system using the full ATX power supply and this time the mobo smelt of smoke. No power or BIOS. I only did this second attempt in order to disable a 'cool and quiet' function that the AsRock bios has built in and enabled by default. The manual says leaving this enabled may cause PSU problems. I suspected this would be the problem that is failing both my Pico PSUs. Unfortunately, I could no longer access bios using the full ATX PSU as the mobo smelt like it got fried.

5. I've initiated a mobo RMA and ordered an Amazon same day delivery of the SAME mobo, make and model.

5. On arrival, I plugged in the throwaway £8 Pico plug and the system booted up instantly on first attempt! So I tried the more expensive branded Pico and it refused to boot.

6. It seems I must have had both a dodgy Pico psu AND a dodgy mobo all at the same time! Why else would none of the power supply units power it on, (except for that one time it did using the full ATX psu but failed on the second attempt)? How else do I explain that using a replacement mobo of the same make and model powers on flawlessly, using the previously tested 'failed' £8 Pico PSU?

Conclusion:

I can't say I have the answers. All I know is that the £8 Pico PSU showed me all the right readings using the multimeter and still failed to power on the first mobo. Replacement mobo powers on just fine using this cheap Pico. The £40 Pico does nothing.

Let me know if I can use this thread to ask you guys for some further help. I could really do with some more guidance. Thanks to 'the hack' for holding my hand so far.
 

Damascus

Master of Cramming
Feb 27, 2018
551
388
Hey, just dropped by your thread and did a quick read through. If you're interested, I have an hdplex 160 + 160 brick that I'm not using that I could send it your way relatively cheap. I know it's not a pico unit, but it's garunteed to be fully functional and it would probably cost about the same as the amount you could get if you return your questionable units.
 

Steedsofwar

Chassis Packer
Original poster
Oct 14, 2018
15
0
Guys, using the streacom F1 Evo case and the stock and cooler for my Ryzen 5 2400g, what temperatures are normal? I have a 40mm noctua intake and a 60mm unbranded but good exhaust set up as my case fans. The fans are set to be quiet for regular use but ramp up to a 100% after exceeding 68 70 Celsius. No overclocks. AsRock b450 gaming ITX motherboard. I have set the bios to enable the one xmp profile it has, to the same 2400mhz speed that the ram is rated for, at 1.2v. It's an auto setting.

My temps are alarmingly high, as compared to my much bigger better specced main PC.

After booting up and letting the temps rise to idle levels, the Ryzen apu sits at 20 degrees Celsius higher than ambient. Sometimes less but twenty higher, most of the time. So now I'm at 24.4c ambient, CPU is at 44°C, doing nothing at all. Even in idle, the temps randomly spike up to over 52°C.

In regular tasks such as browsing or streaming, the temps go well over 64°C regularly, when opening apps like chrome and then sitting at mid to high 50s. Using prime 95 v26.6, the small ftt test sends it through the roof. I turn it off at 90°C, which it reaches in twenty-five seconds. CPUz shows a voltage of 1.39v.

Is the stock voltage levels too high for this Ryzen chip, for CPU AND memory? Is it that my coolers are not good enough in this setup? Is it that such USFF setups are expected to produce these kinds of temps? Or do all three of those apply?
 

Choidebu

"Banned"
Aug 16, 2017
1,198
1,205
With no ventilation above the cpu cooler it is to be expected. Maybe try without the top lid and see the difference, if not much then double check thermal paste and cooler contact.
 
  • Like
Reactions: Steedsofwar