The following information seems to generalize to any of MingHe’s buck/boost converter designs with the 8x2pin daughter board with display and buttons).
There’s a good writeup about this device submitted by a reader. I’m not really a fan of this device as it seems the quality varies widely (and voltage overshoots are never a good thing), so be careful if you get one.
Though most of this info is for the more common MingHe buck or boost converters, I happen to have the MingHe D3806 Buck+Boost converter:

Many reviews are out there for the B3603 and some for the BST400. They all seem to have the same basic functionality on the control board, though the calibration points will differ depending on available output ranges.
My experience with this device seems to be in line with that of other reviewers – namely, it works well with no major issues. However, there are two things to keep in mind:
I’ve noticed that light loads (e.g., 5V < 200 mA) tend to cause momentary voltage overshoots on power down (the lower the load, the more the overshoot)! even a short burst of 12V into a 3.3V or 5V circuit can kill a device, so beware. Might want to add a small (say 2W) load resistor for safety for small loads (if not in CC mode).
As others have noted, if you press and hold “Set”, you enter calibration mode (you’ll see something like “–F1–” on the display). If you didn’t mean to enter cal mode just power it off right now, then back on, to ensure you’re out of that mode! See the EEVBlog review below for how this can otherwise quickly lead to electronic sadness if you accidentally step into the higher voltage cal settings unwittingly.
Notes on the Function Menu settings (AKA the “secret” menu since it’s not exactly obvious):
Hold OK on power-up and wait for the display to cycle -0-, -1-, -2-, -0-… Release OK to toggle and then view that setting. Power down and repeat the process as necessary.
- -0- — y=output on initial power; n=output only after pressing OK (I prefer this)
- -1- — y=show V, A, power, and cumulative amp-hour displays; n=V, A only
- -2- — y=cycle displays by default; n=don’t cycle by default
There seems to be a way to enter quite a few voltage/current presets – check out the manuals for details (seems like this could be prone to accidental selection like calibration is, but YMMV).
Further reading:
- MingHe’s manual for the B3603 (direct)
- MingHe’s manual for the D3806 (via Banggood)
- MingHe’s manual for the BST400 (direct)
- MingHe’s manual for the B3606 (direct)
- A user manual I found for the B3603 (shows how to read the display and exercise the function menu)
- A document with the calibration process
- A thorough text review of the B3603 (Rick Law at EEVBlog)
- A video review of the B3603 (Julian Ilett)
- A video review of the BST400 (Julian Ilett)
- Another, more critical review of the BST400 (Jim Conner)
- Note: I didn’t experience the problem he noted at super-low voltage settings – I suspect it was a buggy old revision of the device firmware
Postscript:
MingHe’s ZXY6005S and ZXY6010S devices are very intriguing! There are older revisions (2012) and newer ones (>= 2013-08) on eBay… I haven’t tried it yet but Julian has it in his round-up video. While I think the D3806 hits the sweet spot for power and flexibility for me, that LCD + rotary encoder is pretty sweet!
Permalink
Did you ever get this to work satisfactory, I Built one in to a bench supply and despite getting an initial very high over voltage 124 vdc! that was the only output.
Permalink
If you got a faulty one and it threw 124V through the output capacitors, it’s surely broken now. I’m not saying I recommend them either – just that mine seemed to work. I’m sorry to hear you got one that failed that badly.
Permalink
Can you control the output digitally via the Rx & Tx inputs on the control module?.
Permalink
I’m not sure… it’s not clear this is even possible without modifying the firmware (and that’s it’s own mystery). Some discussion about this for the B3603 is found at https://github.com/baruch/b3603
Permalink
But.. Is the controller the same as the B3603? I mean, can I download on a D3806 the B3603 firmware found on github?
Permalink
I have no idea… I really doubt it’d work.
Permalink
Hello … I bought one piece from ebay UK i used laptop power supply 19.5v , 4.5A to supply converter, ok the device boost ” step up ” voltage from 19.5 v to above but don’t step down ” not reduce ” voltage !!???? The device is boost and buck , any idea what is problem some one can help… Thanks.
Permalink
Mine worked for that use case but the more I learn about such cheap converters the less I trust them. That said, it ought to work and if it’s not, it may be faulty. Talk to the seller.
Permalink
Thanks , i think a problem with ic XL4016 it’s voltage regulator ic , and it’s not brand device as you say , i reset it to ” factory reset ” but still not work.
Permalink
How have you solved the problem, have replaced XL4016?
Permalink
https://rmprepusb.blogspot.com/2019/11/is-d3806-boostbuck-dc-dc-power-supply.html
I have a 30 Ohm load and still get large 12v spike\surge on the output.
This thing is useless!
Permalink
Thanks for the info! I updated the post to reflect this information.
Permalink
I also found another issue with my D3806 which is really disappointing as it means I can’t use it even if I use a relay on the output to avoid the power-on/power-off pulses:
Try this:
1. Input voltage 12V
2. Output voltage 10V (also tried 20V) it doesn’t seem to matter
3. Connect 52R resistive load (not sure if exact value is important)
4. Switch on 12V source voltage to D3806
5. Press OK button to turn on output
6. Now disconnect load and re-connect it a few times (e.g. dab wire onto load for 1 second a few times. When load is connected look for flashing CC LED. If it starts to flash then leave load connected.
The output suddenly starts to cycle (e.g. if was set to 10V, it cycles up and down 8V to 10V ish. It also starts to draw a lot more current from the 12V source. The CC light can flash on briefly occasionally and the current display on the D3806 will fluctuate.
The output is very unstable. If you press SET to switch off the output and press OK to enable the output, then output voltage is stable.
Does yours do this?
Permalink
I’m not set up to try this right now but that sounds really bad to me. It was one of a few (cheap and clearly dodgy) power supplies I bought along the way, before I ended up getting a proper bench supply (Siglent SPD3303X-E). I really don’t trust this ‘3806 which is why I’ve never done much with it except try it out and back away like I did.
Permalink
Nice Siglent PSU!
I have now bought about different 8 different Chinese power modules hoping to make an ATX-based bench PSU and most have severe problems which render them useless as a general purpose power supply. I have sent back one of them for a full refund and got a partial refund on two others because they don’t work as specified. One vendor even said ‘well, they are really cheap and the cost includes postage from China and to you, plus tax and customs duty – what do you expect!’
Permalink
If they were well-designed I’d be more hopeful, but they tend to use problematic chips and/or knockoffs. I’ve had boards with XLSEMI chips as well as their “LM” knockoffs like the LM2596S. I’d pay to get ones with quality first-sourced regulators… or hell, let me solder my own regulator from TI on it!
Permalink
I would bite the bullet and get an DPS5005 or similar. Those units seem to be pretty well built and thought out.
Permalink
Or, for buck-boost, the DPH3205
Permalink
You might want to take a look at:
https://github.com/rin67630/Drok-Juntek-on-steroids
mingHe =Juntek.
Permalink
That’s pretty cool! Thanks for the tip!
Permalink
You can also have a look here:
https://discord.gg/ZZEs9K3EVw
I have used the electronics of a B3603, removed the Buck converter chip and replaced it by an external buck converter so I can provide double power.
The MingHe B3606 seems to do the same but isn’t available any more…