@jamwaffles , how is your power supply working so far. Is there overheating or issues so far with your PCB circuit. Thank you for your GitHub page. Please post an update. Very keen on doing something similar myself.
Powering a Dell Optiplex Micro with 24v
- Thread starter jamwaffles
- Start date
-
- Tags
- psu tinyminimicro
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
I'm afraid I shelved the project and moved onto other things as soon as I could get to the BIOS screen as proof it was working. The PSU works for sure, but I haven't done any longer-term or higher current draw testing with it yet. The voltage regulator on the board is rated for 5A so the 3.34A max draw from the Dell shouldn't push much heat through it.
Assuming 85% efficiency (worst case from the datasheet), dropping 24V to 19.5V (so 4.5V drop) at 3.34A, that should only be ~2.25W dissipated from the regulator itself which isn't much. I'm not counting any other parts like the inductor, etc.
I'll do my best to update this thread when I get back round to working on the PSU project again
Assuming 85% efficiency (worst case from the datasheet), dropping 24V to 19.5V (so 4.5V drop) at 3.34A, that should only be ~2.25W dissipated from the regulator itself which isn't much. I'm not counting any other parts like the inductor, etc.
I'll do my best to update this thread when I get back round to working on the PSU project again
@jamwaffles thanks so much for the update. Appreciate your reply. I will follow your thread for further updates..
Hey @namus80. I had a spare moment to do a quick stress test on my little power supply board.
Setup:
Optiplex 3050 with i3-7100T, 4GB 2400MHz DDR4, HDMI, Ethernet, USB3 hub (in my monitor) with mouse, keyboard, mic and another hub plugged in. There's an internal SSD and a not-connected-to-an-AP WiFi card on the 3050 I'm testing with.
I have a bench PSU set at 24V input voltage. All current readings are from its display, as input to the little PSU board.
- Power off state: 17mA
- Sitting at Win11 desktop: ~1000mA (but Windows was doing a bunch of crap in the background as usual). The PSU PCB was warm to the touch but not at all hot.
- Cinebench off the MS store: ~1200mA. PSU was almost too hot to touch but I'd only estimate at about ~50C so not close to thermal overload. I don't have a thermometer to get a better reading unfortunately
I think this board is fine for every day use, but it would definitely benefit from a little bit of heat sinking or moving air for better safety margins. I had plans to use a thick thermal pad to glue it to the inside of the chassis. I'll update this thread again if I get round to doing that. Not much room in these little chassis!
Setup:
Optiplex 3050 with i3-7100T, 4GB 2400MHz DDR4, HDMI, Ethernet, USB3 hub (in my monitor) with mouse, keyboard, mic and another hub plugged in. There's an internal SSD and a not-connected-to-an-AP WiFi card on the 3050 I'm testing with.
I have a bench PSU set at 24V input voltage. All current readings are from its display, as input to the little PSU board.
- Power off state: 17mA
- Sitting at Win11 desktop: ~1000mA (but Windows was doing a bunch of crap in the background as usual). The PSU PCB was warm to the touch but not at all hot.
- Cinebench off the MS store: ~1200mA. PSU was almost too hot to touch but I'd only estimate at about ~50C so not close to thermal overload. I don't have a thermometer to get a better reading unfortunately
I think this board is fine for every day use, but it would definitely benefit from a little bit of heat sinking or moving air for better safety margins. I had plans to use a thick thermal pad to glue it to the inside of the chassis. I'll update this thread again if I get round to doing that. Not much room in these little chassis!
@jamwaffles thanks very much for your reply and the stress test. This is amazing, I will try and get some PCBs and parts as per your BOM and recreate it. thanks again for your amazing work.
My field tech rule of thumb for heat is: If you can't let the pad of your finger remain on a heatsink for 3 seconds then its probably more than 60 degrees C. Use your own judgement/metric if you can only touch it for 1-2 seconds. If you burn your finger immediately its probably 100 degrees or more.
Pretty good rule of thumb . I could leave my finger on it probably indefinitely, but it's difficult to gauge the temperature that way as I'm certain my finger was acting as a small liquid cooler - lifting it off and touching the board again, it felt a lot cooler. Either way, it didn't feel like a dangerous (to components) temperature.
. I could leave my finger on it probably indefinitely, but it's difficult to gauge the temperature that way as I'm certain my finger was acting as a small liquid cooler - lifting it off and touching the board again, it felt a lot cooler. Either way, it didn't feel like a dangerous (to components) temperature.