Friday, July 3rd 2026

Thermaltake Intros Dr. Power III Pro Tester for ATX 3.1 PSUs

Thermaltake today introduced the Dr. Power III Pro (model: AC-083-OO1NAN-A1), a handy tool to test modern ATX 3.1 power supplies. This was first showcased by Thermaltake at the 2026 CES. The tester lets you check voltages and stability of the various connectors of an ATX 3.1 PSU, including a 600 W-capable 12V-2x6 connector, a 24-pin ATX, a 6+2 pin PCIe power, a 4+4 pin EPS, SATA power, and 4-pin Molex. The tester lets you check if the various voltage domains are within safe ranges, so you know if your PSU is broken. An LED segment display puts out measurements for the various domains, while a buzzer sounds alarm if something is off-range. Available from July 9th, the Thermaltake Dr. Power III Pro tester is priced at $49.99.
Add your own comment

18 Comments on Thermaltake Intros Dr. Power III Pro Tester for ATX 3.1 PSUs

#1
Chaitanya
Have used 1st version long back. Its a useful tool and its decently priced..
Posted on Reply
#2
KLMR
That is a simple micro, the connectors and a display. But at 50$ its decently priced for its functionality and at reach of anyone. I was expecting 150 at least knowing would be an uberinflated price. If not, look for the uberexpensive derbauer 12V-2x6 prices. Nonsense for even a simpler BOM and software.
Kudos for TT for pricing the device at a profitable but decent price!
Posted on Reply
#3
outlw
This does look like a nice tester, and the price is quite reasonable even!

But... it does not put a load on the PSU, does it?
So, any issues that the PSU has only when loaded will not be found?

I guess it would still be great for quickly confirming if a PSU is obviously defective, but could not conclusively rule out the PSU when troubleshooting 🤔
Posted on Reply
#4
huggi
I've got the non-Pro version of this and the only difference I can find is that this has a colour LCD screen. It also costs an extra AU$30 locally ($89 vs $109) which seems a little steep for a screen swap.
outlw6669But... it does not put a load on the PSU, does it?
So, any issues that the PSU has only when loaded will not be found?

I guess it would still be great for quickly confirming if a PSU is obviously defective, but could not conclusively rule out the PSU when troubleshooting 🤔
Yeah, not a load tester. It only checks that the pinouts and the voltages are correct but that's basically all most people would need. A proper load tester would be a couple hundred dollars I'd think.
Posted on Reply
#5
_roman_
I miss certain parameters like multimeter have, TRUE RMS and the accuracy statement. If there are none, it is a toy with random numbers.

www.thermaltake.com/dr-power-iii.html

something does not add up, release date 2024-02-29 for the manual

is thermaltake a low quality brand?

url did not open, non in normal window or private window, i had to use the shell

wget https://file.thermaltake.com/file/qig/Dr.PowerIII_Manual.pdf
--2026-07-03 15:20:00-- https://file.thermaltake.com/file/qig/Dr.PowerIII_Manual.pdf
Resolving file.thermaltake.com (file.thermaltake.com)... 208.109.246.156
Connecting to file.thermaltake.com (file.thermaltake.com)|208.109.246.156|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 483206 (472K) [application/pdf]
Saving to: 'Dr.PowerIII_Manual.pdf'

Dr.PowerIII_Manual.pdf 100%[===========================================================================================>] 471.88K 133KB/s in 3.5s

2026-07-03 15:20:06 (133 KB/s) - 'Dr.PowerIII_Manual.pdf' saved [483206/483206]

It is a toy.

no accuracy, no digits, and i would not call it a measurement device



A faulty psu can be out of range of those values. And is the psu really out of range when there is no given accuracy like any multimeter has?
Posted on Reply
#6
Destined
This a crazy world we live in!!! Nvidia created the problem, they should solve it....
This tool should be bought by companies that sells power supplies. They should test them before selling them....
if i suspect my psu is faulty, I RMA that shit. Why add insult to injury.
idk, correct me if I am wrong :/
Posted on Reply
#7
Snoop05
DestinedThis a crazy world we live in!!! Nvidia created the problem, they should solve it....
This tool should be bought by companies that sells power supplies. They should test them before selling them....
if i suspect my psu is faulty, I RMA that shit. Why add insult to injury.
idk, correct me if I am wrong :/
and what does nvidia have to do with this?
Posted on Reply
#8
TheLostSwede
News Editor
_roman_I miss certain parameters like multimeter have, TRUE RMS and the accuracy statement. If there are none, it is a toy with random numbers.

www.thermaltake.com/dr-power-iii.html

something does not add up, release date 2024-02-29 for the manual

is thermaltake a low quality brand?

url did not open, non in normal window or private window, i had to use the shell

wget https://file.thermaltake.com/file/qig/Dr.PowerIII_Manual.pdf
--2026-07-03 15:20:00-- https://file.thermaltake.com/file/qig/Dr.PowerIII_Manual.pdf
Resolving file.thermaltake.com (file.thermaltake.com)... 208.109.246.156
Connecting to file.thermaltake.com (file.thermaltake.com)|208.109.246.156|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 483206 (472K) [application/pdf]
Saving to: 'Dr.PowerIII_Manual.pdf'

Dr.PowerIII_Manual.pdf 100%[===========================================================================================>] 471.88K 133KB/s in 3.5s

2026-07-03 15:20:06 (133 KB/s) - 'Dr.PowerIII_Manual.pdf' saved [483206/483206]

It is a toy.

no accuracy, no digits, and i would not call it a measurement device



A faulty psu can be out of range of those values. And is the psu really out of range when there is no given accuracy like any multimeter has?
That's the non Pro version.

And yes, Thermaltake is considered a lower quality brand by many, but they have the odd higher-quality products too.
Posted on Reply
#9
Chaitanya
_roman_I miss certain parameters like multimeter have, TRUE RMS and the accuracy statement. If there are none, it is a toy with random numbers.

www.thermaltake.com/dr-power-iii.html

something does not add up, release date 2024-02-29 for the manual

is thermaltake a low quality brand?

url did not open, non in normal window or private window, i had to use the shell

wget https://file.thermaltake.com/file/qig/Dr.PowerIII_Manual.pdf
--2026-07-03 15:20:00-- https://file.thermaltake.com/file/qig/Dr.PowerIII_Manual.pdf
Resolving file.thermaltake.com (file.thermaltake.com)... 208.109.246.156
Connecting to file.thermaltake.com (file.thermaltake.com)|208.109.246.156|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 483206 (472K) [application/pdf]
Saving to: 'Dr.PowerIII_Manual.pdf'

Dr.PowerIII_Manual.pdf 100%[===========================================================================================>] 471.88K 133KB/s in 3.5s

2026-07-03 15:20:06 (133 KB/s) - 'Dr.PowerIII_Manual.pdf' saved [483206/483206]

It is a toy.

no accuracy, no digits, and i would not call it a measurement device



A faulty psu can be out of range of those values. And is the psu really out of range when there is no given accuracy like any multimeter has?
Here is the link to correct product:
www.thermaltake.com/dr-power-iii-pro.html
Posted on Reply
#10
r00tcz
While useful, checking PSU voltages with completely no load is quite uninteresting, just basic PASS/FAIL if PSU is likely to be OK. It would be far more useful to have say XT90 load connector (where you can put something like couple of car lightbulbs or cooled resistors etc.) and internal MOSFETs to put that load on different voltage rails to see what happens (measuring transients and voltage drop and other things). Maybe this is idea for future Dr. Power IV Pro Mega Ultimate Edition.
Posted on Reply
#11
Dr. Dro
DestinedThis a crazy world we live in!!! Nvidia created the problem, they should solve it....
This tool should be bought by companies that sells power supplies. They should test them before selling them....
if i suspect my psu is faulty, I RMA that shit. Why add insult to injury.
idk, correct me if I am wrong :/
You're wrong, this has nothing to do with Nvidia, Intel, the PCI-SIG or the 12VHPWR connector. Dr. Power products have been available for decades at this point. It's simply a multimeter designed to test if a power supply is working. Proper lab-grade load testers like the SunMoon SM-8800 units cost thousands of dollars and aren't available for purchase by the average customer, even if they had the money.
Posted on Reply
#12
R-T-B
Snoop05and what does nvidia have to do with this?
He's trying to transform this into yet another 12VHPWR discussion.
Posted on Reply
#13
Wirko
r00tczWhile useful, checking PSU voltages with completely no load is quite uninteresting, just basic PASS/FAIL if PSU is likely to be OK. It would be far more useful to have say XT90 load connector (where you can put something like couple of car lightbulbs or cooled resistors etc.) and internal MOSFETs to put that load on different voltage rails to see what happens (measuring transients and voltage drop and other things). Maybe this is idea for future Dr. Power IV Pro Mega Ultimate Edition.
One thing you can do is buy splitters for all the power cables (or at least the 24-pin motherboard cable) and have this instrument measure voltages while the PC components are powered on and under load. Not nearly ideal because the load is never constant but much better than testing on idle.
_roman_I miss certain parameters like multimeter have, TRUE RMS and the accuracy statement.
You want to measure true RMS value of a DC voltage? It's not impossible, and some multimeters with a DC+AC RMS function can do just that. But why?
Posted on Reply
#14
Jism
Weird thing. You obviously want to load test a PSU with for example, dated, old capacitors to measure any fluctuations big enough.
Posted on Reply
#15
duynguyenle
DestinedThis a crazy world we live in!!! Nvidia created the problem, they should solve it....
This tool should be bought by companies that sells power supplies. They should test them before selling them....
if i suspect my psu is faulty, I RMA that shit. Why add insult to injury.
idk, correct me if I am wrong :/
Power supply OEMs do, in fact, test every PSU before packaging and selling them. Production testing of PSUs also involves much more sophisticated load testers than the Thermaltake gizmo in this article, which is a glorified digital multimeter that just displays zero-load voltage on each rail of the PSU. If you know how to use a multimeter probe, you do not need this product.

I am not sure what you're even on about here. If you think your PSU is faulty, it IS recommended to RMA them.
Posted on Reply
#16
twin_savage
I think some here are missing the point of this tool. It's to sanity check the cables you have.
It would have prevented this:
www.tomshardware.com/pc-components/power-supplies/evga-power-supply-allegedly-blows-up-22tb-of-customer-storage-revised-model-from-rma-had-a-different-pin-layout-and-killed-all-sata-powered-devices

Modular PSU pinouts aren't standardized and often the connectors aren't uniquely keyed among manufactures so it is entirely possible to fry your devices using cables that plug in just fine to your PSU.
This tool is also useful for checking that you have correctly constructed your own cables.
Posted on Reply
#17
_roman_
TheLostSwedeAnd yes, Thermaltake is considered a lower quality brand by many, but they have the odd higher-quality products too.
Thanks. Well it seems i also failed on reading the full product description and pulled up the wrong product.

Companies should use more clear names and stop using the same name. I did not do it on purpose, this time I really got the wrong serach result from taht homepage and the wrong data.

well teh manual again did not open in a browser. I will not share the measurement table as it is 1 vs 1 the same as the screenshot above.

I do not see any given accuracy + digit error on the pro also

Note: just you see what file i looked at, url.
roman@Wheat_Nas ~/Downloads $ wget https://file.thermaltake.com/file/qig/Dr.%20Power_III_Pro_Manual.pdf
--2026-07-06 19:36:57-- https://file.thermaltake.com/file/qig/Dr.%20Power_III_Pro_Manual.pdf
Resolving file.thermaltake.com (file.thermaltake.com)... 208.109.246.156
Connecting to file.thermaltake.com (file.thermaltake.com)|208.109.246.156|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 389794 (381K) [application/pdf]
Saving to: 'Dr. Power_III_Pro_Manual.pdf'

Dr. Power_III_Pro_Manual.pdf 100%[===========================================================================================>] 380.66K 125KB/s in 3.0s

2026-07-06 19:37:01 (125 KB/s) - 'Dr. Power_III_Pro_Manual.pdf' saved [389794/389794]

Verdict: Toy. Not a measurement device because it lacks certain expectation whcih are common even for 10€ supermarket multimeters which are unsafe to use. First rule which I was told in school around 1996. What are you measuring and where are you measuring and how are you measuring. Is it a valid measurement. At that time we would blow fuses which could be triggered again mechanically when we set wrong measurement ranges. It was important to know beforehand what measurement range, etc. - else the fuse would blow and everyone would notice it. Analogue measurement devices are a technological high end piece.
WirkoYou want to measure true RMS value of a DC voltage? It's not impossible, and some multimeters with a DC+AC RMS function can do just that. But why?
Any multimeter which is not junk has since 1990 a label on it TRUE RMS.

You are right on your argurment. Anyway why should i bother if it has or not has true rms, when any of my devices has it?

www.fluke.com/en-us/learn/blog/electrical/what-is-true-rms

When I want a bad hacky product. Take a Pic16f84A + some resistors + ADC ports. This product here does not something else

Where to measure true-RMS

The need for true-RMS meters has grown as the possibility of non-sinusoidal waves in circuits has greatly increased in recent years. Some examples:
  • Variable-speed motor drives
  • Electronic ballasts
  • Computers
  • HVAC
  • Solid-state environments
Feel free to elaborate.

When i see computers mentioned on the fluke website I just realised how many samples are taken?

I think you gave me even more fuel about this being a bad product. How many samples are taken? Is it a valid measurement? en.wikipedia.org/wiki/Nyquist–Shannon_sampling_theorem

I am aware of my true rms current clamp meter is a valid measurement device for checking if the generator, alternator and battery is fine on my diesel car. but not for computer etc because of the low sample size.

We do not measure a lightbulb where the current and voltage hardly changes. Computer changes a lot the currents. That was a bad argument with your true rms. Back to square one, what is a valid measurement device and waht is a toy? Regardless of your true-rms statement which was in fact true and I have forgotten about the AC parameter of it. Is it AC, DC or mixed voltage and current in computer? which wave form?

Back to square one, what is a valid measurement device and waht is a toy? A good product does not reveal more question but answer those in less than 1/4th a4 page in the manual on the page 1.

I did bother with the motherboard and how it seems to be measured there. the scope for the motherboard just is to know if the voltage exists in the given range or not. Therefore the low sample size and the 4% resistor values are enough to determine for the novuton chip all those voltages to initalise the motherboard.

Whatever P.G is in the manual, a time range of 0.1 to 0.5 Seconds can not really determine how stable the voltage is because of too high measurement range.
See news piece:
The tester lets you check voltages and stability of the various connectors ....
Min, Max, Average for all measurements is a given with +/- 1% +2 digits. (that is a cheap measurement range definition)
Regardless of the waveform, what sample size should be taken, is something which guys who are paid and know more details should decide. I would start at those resolution, which e.g. ripple are usually for psu and than add a factor 10 in measurement ranges.

It would not be a bad product if the user can see in a way that his psu is out of spec. therefore ATX 3.1, ATX 3.0 and the older spec switches etc. should be included to test for that.

A nice product, just badly executed, especially with the claim to check for voltages and stability.
Posted on Reply
#18
Wirko
@_roman_
Here we're talking about measurement of 12 V DC and other DC voltages coming out of the PSU. Meanwhile, a true RMS voltmeter measures the AC component (the ripple voltage) or, in some cases, DC+AC (but I don't see the usefulness of this on a computer PSU).
So you want to know the ripple voltage? Good, but what is the sufficient bandwidth? Probably above the PSU switching frequency, while a cheap RMS voltmeter only reaches about 1 kHz, so you need an AC-coupled digital oscilloscope. And even then, is it really the RMS value that matters most, or is it peak-to-peak?

When testing a PC PSU, I'd rather have an instrument that captures voltage dips at a high enough sampling rate. Those dips can potentially make the system unstable. I have no idea what a high enough sampling rate would be, perhaps between 200 and 2000 Hz?

However, I agree that a $50 voltmeter should have its accuracy written in the specifications, just like most cheap multimeters from Aliexpress do (such as Zotek/Zoyi, great little tools, I have three).
Posted on Reply
Add your own comment
Nov 18th, 2026 14:30 PST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts