DP100 PD-Power Supply 100Watt [0-30.00V / 0-5.000A) [∼50$]

[DaneLaw]

Has anybody tried this small power supply that resamples the Miniware MDP series quite a lot. .?

It's tiny, at around 10cm in length 1½cm high, and 6cm deep, but looks quite feature-rich, with 4mm banana jacks fitted with corresponding red/blue ring lights when power is active, TypeC with fastcharge protocols like PD3, USB-A for peripherals like mouse-control, also seems to have a PC interface.
It could be a good fit with these modern high-wattage PD3.0 and PD3.1 power banks as a small mobile solution.
In transit and should arrive in the next week - A small switching alternative to the two bulky old linear Korad PSUs I got, which are cumbersome to pull out, and set up with PC control.

Features
**
DP100 CNC power supply is a high-performance CNC adjustable DC power supply
Power supply mode Support PD/QC fast charging source, universal DC power adapter, charging treasure
Support 10 groups of preset output, directly call out
The power-on curve is smooth and stable without overshoot
Hardware constant voltage constant current circuit, intelligent anti-burning, constant voltage constant current mode automatic switching
The input supports reverse protection and undervoltage protection
The output supports overvoltage protection, overcurrent protection, overload protection, overheat protection and reverse protection
USB-A interface slave mode (USBD) supports communication with the host computer
Use 0.96 inch 160*80 HD IPS screen to display and control panel humanized Angle design, easy to operate
Size: 10.4cm*6.2cm*1.72cm
Input voltage: DC5V~32V
Output voltage: 00.00V~30.00V, resolution 0.01V continuously adjustable
Output current: 0.000A~5.000A, resolution 0.001A continuously adjustable
Output power: ≤100W
Conversion efficiency: ≤97%
Input interface: Type-C
Output interface: 4.0mm banana head interface
Fast charge protocol: PD/QC
Screen: 0.96TFT (160*80Pixel, IPS)
**







PC software that seems to have english support..



A look inside (+4.10min)

[YTusername]

Hi,

Any chance you tried it?

I'm looking to order one too.

There is a review video (Japanese / subtitle available). It seems very good for the price.

[youtube][/youtube]

[DaneLaw]

- Any chance you tried it?

They first arrived today, and its orders from Singles Day nov-purchase in one of Aliexpress's so-called "fast combined-shipment" where they will try to bundle it with numerous of your other orders from those weeks, but December month is a heavy shipping month, and the delay is likely down to our local PostNord that will gladly let your China parcel linger locally for weeks' before they will deliver it.

So I haven't actually tried it to any real extent, only a brief visual look today on one of them... its very Tiny..
The possible add-on "100W TypeC" cable comes in a separate box  and the embedded chip in the cable reads it as "20V / 5A - USB3.2 gen2"
Overall it's a nice 100W 150cm TypeC-to-TypeC silicone cable with good lock-indents.

The black & red casing also looks pretty good with a riffled ingrowed surface-texture, though it has quite a fabrication smell to it, so it's on the balcony sprayed with some Rodalon so hopefully the smell will be a saga.
The incl. 4mm banana to crocodile clips are goldplated in both ends with silicone wires.(16awg 200 degrees rated)
Here at first glance, the DP100 unit looks good - and seems to match the description, the knurled metal roller, sadly aint clickable so you validate/enter with the front buttons.
The incl. manual does cover English, foldable with one side CN and the other ENG.(10 pages)

// with the price in mind, it looks like a nice unit (sofar) but also precuriously cheap, so still waiting to see where the cat is buried
(vers. 1.4_1.4)

[YTusername]

Nice shoots! Thanks

// with the price in mind, it looks like a nice unit (sofar) but also precuriously cheap, so still waiting to see where the cat is buried
(vers. 1.4_1.4)

I think the price is reasonable. I know it is generally cheap, but the RD60xx serise has more features and prooved performance with arround 64-112$ based on PSU and case selection.

[DaneLaw]

The RuiDeng units, It's another fish, nobody would pursue Atomics DP100 or Miniware MDP-series, for PSU purposes, if size benefits weren't critical, then you would go with a full fledge bench PSU, if mobility & size weren't a crucial factor.

Sure' you can get a Ruideng control PSU panel for relatively small funds, but its two very different-sized products in build and design, the DP100 its tiny, to the point that many would see that tinyness as a con, if the tiny size wasn't something they needed.
Here in my taxed northern region, RuiDengs RDxxxx products are from. 130USD for what looks to be a barebone RD6012 frontboard (no wifi) and up, incl. enclosure-backcase it's 170 to 180USD and then you can add a PSU.. https://aliexpress.com/item/1005003472332331.html?

Already have two linear Korads-P units http://tinyurl.com/28upwfre and they are bulky as, (just like linear PSUs tend to be for the linear benefits) and certainly not mobile, and it's the small form factor in fx DP100 that is so easy to just pull out on your desk or outside, and attach with a high wattage power bank, that appeals to me, not least to running high powered 9mm laser diodes, as im fascinated by light, and particularly coherent wavelengths, where many of the high power lazing diodes nowadays are quite resistant to switching PSUs.

DP100 48 USD (incl. 25% EU-VAT and delivery) but it is a sale price (BF/SD sale) where Aliexpress eats a good chunk of the price & takes it down a nudge http://tinyurl.com/mupskyju

Been very close to take the plunge on the Miniware variant since it came out "MDP-X" fx 905 but the more I looked into it, the more limited it seemed hence the choices Miniware had gone with, though Miniware did try to counter some of the shortcomings with later revisions, not least protocols.
The fanless MDP P905 (90W) which in the standalone version costs around (160USD) though P905 discontinued now, for a beefier variant P906 (220USD / Eleshop) which is now also holding a tiny hysteric fan (not dat creazy about that) likely a must in such a small form factor, being able to reach 10A.

I'm not familiar with other alternatives, then the Miniware MDP PXXX variant and then the DP100 from Punctual Atomic, all other units seem to be in another adj. PSU category, also RuiDeng units.

So far I do like the protocol approach in the DP100, as you have quite big control over the power delivery/PD and Qualcomm protocols when supplying power.
PD (to a type C PD socket) you can cycle at will in the DP100 menu, 9v, 12v, 15v, 20v, and your power bank or PSU will adjust accordingly on the fly.
Qualcomm quick charge (when using power from a standard USB-A port) you can cycle between 9v and 12v.

Tried with an Anker737 power bank (PowerCore24). it got 2x TypeC (PD3.1) and 1 USB-A (QC support).. the 2 TypeC sockets would cycle 9v, 15v, 20v (this 737model PB doesn't support 12v PD) and running it from the USB-A port would let me choose between 9v and 12v (QC)

You can always just supply ordinary non-protocol power, to get above 20v, as the unit doesn't support PD3.1 which opens up for fx 28v/5A (140W) DP100 got PD3.0 as far as I can tell, that maxes out at 5A 20v (100w)
The Anker737 power bank supports PD3.1 and 140watt of each of the 2x TypeC sockets, so I attached a PD3.1 28V 5A decoy/trigger cable, so the power bank triggers 28v-power delivery out (PD), and that works fine, gives you 28v input on the DP100, - the DP100 is solely buck'ed, so your max will be around 1v lower than your input.

If you dont wanna rely on protocols, you can obviously also do that, fx from a barrel DC, as it comes with a DC-barrel to TypeC adapter that does seem quite beefy hence the metal in the small cable part, likely rated for the 100W.

Tested to see if it overshooted and to what extent...tested when turning the device on & off, and tested when turning the output on & off..
could look like there is some when turning the output off (peaks with +0.4v from 15v output, a brief spike to 15.4v before dropping), and when turning the DP100 on manually with the left button, the same spikes with around +0.4v to 0.400mv before centering. (not the actual output, solely the device on) but there are quite many variables, as you can also set it to turn on aut when connecting the power bank (was using the 737 Powercore24)...

Will try to hook one of these on the input from the power bank to see what it gets in.

[Dexter2]

The positive:
As others have mentioned the power supply is exceptional small, the images on the internet do not do it justice, only when you hold it in your hand you grasp how small it is. The build quality is exceptional, also the accessories are of high quality. The device output voltage is spot on. It agrees in a range of few 100uV to a precision calibrated 7 1/2 DMM. Everything is excellent up to this point. It's also power efficient. When loaded with full 5A @12V it only heats up to 40°C.

Neutral:
User interface navigation is a bit non logical, but when you get the hang of it its OK. I was not able to get it to trigger a PD3.1 USB-C power supply to output 28V. So it seems that up to now only up to 20V input is supported over USBC PD protocol. This is a shame as having the power supply to output 24V is beneficial for lot of projects, as most of the industrial electronics is powered by 24V. This is something to consider and i hope could be fixed with a firmware update.


Negative:
From measurements on my received unit i would not recommend using it for sensitive electronics with an external USBC PD power supply. When output is disabled, there are output spikes that go to about -8V. This happens if there is a light load connected that does not have blocking diodes. I tested with two USB-C power supplies, a PD3.0 one and a PD3.1 one. This probably could be fixed with firmware as i did not observe this problem when powered from a standard external 30V power supply. But having such a small laboratory power supply the number one usage case is you pair it with a small GaN USB-C PD power supply.

12V Output ON (Light load with LED + Resistor) using USBC PD3.1 power supply


12V Output OFF (Light load with LED + Resistor) using USBC PD3.1 power supply

Will test more around this problem in the next days. Has anybody seen this same issue with their PD100?

12V Output OFF (Light load with LED + Resistor) using laboratory power supply as source set to 30V.

[YTusername]

User interface navigation is a bit non logical, but when you get the hang of it its OK. I was not able to get it to trigger a PD3.1 USB-C power supply to output 28V. So it seems that up to now only up to 20V input is supported over USBC PD protocol. This is a shame as having the power supply to output 24V is beneficial for lot of projects, as most of the industrial electronics is powered by 24V. This is something to consider and i hope could be fixed with a firmware update.

This confirms what DaneLaw reported too. PD3.1 is not supported.

Negative:
From measurements on my received unit i would not recommend using it for sensitive electronics with an external USBC PD power supply. When output is disabled, there are output spikes that go to about -8V. This happens if there is a light load connected that does not have blocking diodes. I tested with two USB-C power supplies, a PD3.0 one and a PD3.1 one. This probably could be fixed with firmware as i did not observe this problem when powered from a standard external 30V power supply. But having such a small laboratory power supply the number one usage case is you pair it with a small GaN USB-C PD power supply.

Does it have negative spikes behavior when being fed a 5V USB power supply?

[Dexter2]

I continued my investigation with no positive results.

I tried various USB-C cables and 5 different USBC power supplies ranging from 35W to 140W. Even tried a standard 5V USB power supply the result was the same. I even tried to measure the output with an Keysight InfiniVision osciloscope if there was some strange ground loop interaction with my R&S scope but the result was the same.

DP100 output when its disabled over the on screen ON/OFF button. Power source was a USB 5V power supply
No load on output!



DP100 output when its disabled over the on screen ON/OFF button. Power source was a USBC PD3.0 power supply
No load on output!


DP100 output when its disabled over the on screen ON/OFF button. Power source was a USBC PD3.0 power supply
10K as load


DP100 output when its disabled over the on screen ON/OFF button. Power source was a USBC PD3.0 power supply
470R as load


DP100 output when its disabled over the on screen ON/OFF button. Power source was a Keysight low noise loaboaratory PSU
No Load


So what this shows is that the DP100 sadly has some internal problems when the output is disabled, and it starts to output noise that can be in almost -8V range with no or light load on the output when using USBC as power source for it. Even when its powered with a super low noise power supply some noise can be observed when you disable the output. Using a load resistor always connected to the output solves the problem to some degree but the output current measurement is than not correctly displayed on the screen anymore. So i would say a shame as otherwise its quite precise and low noise as long you do not disable the output... So for high current loads like motors, battery charging or loads that draw >100mA its a fine small power supply. But for sensitive low power electronics it has problems to the point you can damage the circuit you are powering with it when using USB power as source...

[Phil1977]

Just my first thought: Why not add a diode antiparallel to the output? If I see the traces below, it should prevent the worst part of the behaviour.

My second thought. Why is this diode not integrated in the design? Cost is neglectible but it can prevent a lot of damage when the power suppy is accidentaly backfeeded.

[Dexter2]

I took it apart and it was immediately apparent why the strange behavior when you disable the output of DP100.

There is quite a bit of capacitance on the output 1600uF in total to be exact (3x220uF + 2x470uF). So disabling the output BUCK regulator would take some time for the capacitors to discharge, and probably the engineers wanted the output to be disable instantly so they added and MOSFET on the output. But to save money they used n N-Type MOSFET on GND.




This creates issues when you have ground referenced electronics, witch most of electronics is even if its battery powered you ground it during development when you connect an oscilloscope, function generator or any measurement equipment... And then the DP100 performs as seen on my measurements when you disable the output. All AC-DC power supplies are capacitive coupled and designed in a way that the output is clean when output negative is connected to ground or protective earth. ALIENTEK PD100 disconnects this connection so depending on the power supply type you get large mains frequency noise on the output.

In my opinion the only use case for this power supply to work reliable is to use it with battery power like a power bank without USB PC connection. If connected to a PC for control over the supplied software, it will again have ground related problems... 

It seams their engineers thought of this problem and wanted a capacitor on the output to couple this AC mains noise during output OFF, but there was mechanically no room for it when assembled, as the output terminals are in the way so this got removed in the final design. So i now added a small 4.7uF MLLC one. Its enough to reduce the output noise to about 200mVpp when output is disabled.



So with this capacitor added Batteries can still be charged and will not discharge when output OFF, there is still polarity protection at least as long the output is OFF and the main current path is interrupted not to discharge the whole 1600uF into the attached load when Power OFF is pressed. But its only a quick Fix as still there is a possibility of unwanted current flow into your attached load. They really should interrupt the positive terminal.

[DaneLaw]

..

Very interesting find, Dexter2..I have two of these from around half a year back and didn't notice any fundamental issues when I got them and briefly tested them.. biggest con was the second unit and roller were not as smooth as the first in its clickable hardware feedback.. it feels like its spring leaf that are clicking while rolling are not aligned correctly, and simply too far ahead.


But all my use-cases with them, are from power banks and that could be why I'm not seeing it, if it's only when you hooked through your AC grid with an AC-DC PD wall-outlet that your getting these negative spikes..?
All your negative spikes were obtained when running it from an AC to DC PSU PD Wall outlet.?

They are quite input willing..Even the crude/standard 5V USB-A with a single 18650 will work, and also PD (power delivery) 5v, 9v, 12, 15v, 20v, but could also look like Qualcomm (QC) 5v, 9v, 12v are supported.
You can also run them from 28V / 5A PD3.1 power banks, like I did above, but you need to trigger before the DP100, as they don't have PD3.1 (28v/5A) trigger naively (not many items have that protocol yet, mostly higher-end laptops from fx Apple, but the PD3.1 chip doesn't cost a lot if you wanna do it yourself on a cable... its like 1½ or 2 bucks with a TypeC female connected... got a couple of them.. it also support 48v, but there are no items that sofar takes advantage of PD 48v.. so its mostly the 28v/5A that are activated from scratch in that chipboard.. the up to 48v needs that you unsolder a made bridge for another to activate that on the PD3.1 board.

[Dexter2]

But all my use-cases with them, are from power banks and that could be why I'm not seeing it, if it's only when you hooked through your AC grid with an AC-DC PD wall-outlet that your getting these negative spikes..?
All your negative spikes were obtained when running it from an AC to DC PSU PD Wall outlet.?

Yes negative spikes are only there if powered from AC PD wall outlet and then the load connected to  the DP100 is light and ground referenced. For me this is the most common use case i wanted to use the PSU in my lab. If you want to repeat my findings you will see the spikes if you just connect your DP100 to a AC PD wall outlet and connect a standard table oscilloscope to the output of the DP100. The spikes appear when you disable the output.

When the DP100 is powered with a battery bank and not connected to a PC there should be no issues. Also if you power a load that has a lot of capacitance itself or has high power draw (more than few 10mA) there also should be no issues.

Its a nice small power supply. If they only would make it to internally disconnect the positive rail and natively support PD3.1 it would be perfect 

[DaneLaw]

Video with tear down..
Dave is quite impressed by this tiny little fellow, that a viewer shipped him in the mail back.
He seems pretty bamboozled that they can sell a product like this at these prices.
Impressive value for the money but also emphasize that if you running it from a third-party AC wall PSU you can transfer noise etc from those devices..
- as there are limits for how much filtering that gonna be in this tiny fellow to clean what is coming in..
but its performance is quite impressive, not least with the price in mind.

[Phil1977]

Thanks a lot for the output schematics!

The Low-side-switch seems really annoying. Do I get it right that if you use this unit e.g. with a vehicle grid source and its output is switched off then power can still flow "around" the low-side-FET if the connected load has any additional mass-connection? That sounds like a problem source directly from hell 

Another issue I see with this setup that is common with many other small adjustable SMPS: There´s a lot of output capacitance that seems to be *behind* the current regulation. This easily kills e.g. LEDs or even Laser Diodes if you ever connect the load to the switched on PS.

Anyhow, it seems to be a sexy device as long as you know about its limitations...

[Hydron]

I'm wondering if the low side switch thing could be worked around with a minor hack - if the positive terminal is fused by that smd component on the bottom of the board (and isn't connected to anything else), could the fuse be replaced by a high side MOSFET driven by the same signal as the low side one? Would need some level shifting but there's likely an integrated part available to make it simple, and add back current limiting too if the fuse isn't replaced.

Kinda regretting not grabbing one during the AliExpress anniversary sale - this issue put me off but that was before the later posts with explanation of the cause and Dave's review.

[Dexter2]

I was thinking the same to make a small piggyback PCB that would connect to the Fuse terminals and use the control signal for the low side MOSFET to control an high side P-Channel MOSFET on positive terminal. There are no connections after the fuse as far i have seen. But this solution would need as you mentioned level shifting and a negative voltage generator to proper open the P-Channel MOSFET also when the output voltage is set below a few V. Not a problem to make but there is not a lot of space available for the piggy-back... Will see if i find a integrated solution that would be small enough to fit. If i find something appropriate i will share my findings.

As for the current regulation, yes it's slow no way to connect a LED with output ON without a bang  But this is true for most switching power supplies that do not have linear post regulation. For this price i thing its an OK tradeoff... It was a better choice to make it low noise but slow than to make it fast and noisy. But probably they could have reduced the output capacitance to at least half with same noise performance by using an LC filter on output and not only capacitors as now...

[Dexter2]

First idea... Will see if it fits...

[Hydron]

Something like this could simplify the circuit (only needs a couple of 100n caps and N-ch MOSFET), though it would need 5V for Vcc rather than 3V3:
https://www.analog.com/media/en/technical-documentation/data-sheets/ltc7004.pdf

Is the USB-A port on the unit always on? That could provide a 5V supply.

There may of course be other better options out there (something that's easier to dead-bug than a 10 pin MSOP and runs on 3V3 would be nice).

[Dexter2]

Nice chip this LTC7004. Wil investigate if there is an internal 5V rail that is always on...

Update: Internally there is a 5V rail that is always present.

[DaneLaw]

You can also control it from a browser fx Chrome.
As it supports WebHID
Video below
Works from a browser link, no program installment.. its from the same fellow that made the open-source input device "Smartknob" with millions of views, that went viral some time back.
He seems to be quite fond of his DP100 purchase.. impressive value for the price.

Open source SmartKnob..


WebHID DP100
Works like a charm.. no programs needed.. just open this link and connect.
https://scottbez1.github.io/webdp100/

Relevant github
GitHub: https://github.com/scottbez1/webdp100

[DaneLaw]

Dynamic Acoustic Vibration Effect, or just DAVE for short (patents filed 1-4-24)
Riding the subharmonics.

[Dexter2]

Designed the update PCB. Should fit nicely to the output capacitors and terminals and will need only two additional small wires. I will upload Gerber's after i test it, if someone else wants to update their DP100. At the end i decided for the more passive solution as its lot cheaper then the LTC7004.

[Phil1977]

Maybe a stupid question, I´ll anyhow speak it out: May anyone imagine putting a small 5A capable relais into the output? Combination of the integrated Low-Side-FET-Switch with a High-Side-Relais feels like a no-brainer-solution for me, but is there by any way enough room for it? I´ll receive my own unit in a around 2 weeks...

[Dexter2]

Maybe a stupid question, I´ll anyhow speak it out: May anyone imagine putting a small 5A capable relais into the output? Combination of the integrated Low-Side-FET-Switch with a High-Side-Relais feels like a no-brainer-solution for me, but is there by any way enough room for it? I´ll receive my own unit in a around 2 weeks...

There may be enough room for a super miniature relay on the top of the PCB where the caps and the main inductor is if its 10mm high it maybe would fit. But anyhow it would be hard to assemble as you would need more wires that go to the other side and would also need to cut the negative trace from the on board MOSFET to the output, probably better to use an extra SOT-23 Mosfet for relay control. It would not be a simple clean solution but doable without the extra mini PCB.

[kloetpatra]

I think these are the smallest you can get for 5A: 10 * 6 * 5.65 mm (l*w*h)
https://www.mouser.de/ProductDetail/TE-Connectivity-PB/IM03DGR?qs=g%252BEszo6zu8MmlSFij8F0cw%3D%3D