Help me Repair an DPS5005

[BeetleJuice]

The story : this unit was purchased a year ago for 35$. i used a transformer 220->36 VAC 300VA , a 20A bridge rectifier and a 1000uf/63V cap 63V (ordered later an 20.000uf/63V).

the unit worked perfect BUT when i pluged the in/out connector i saw a spark between the connector and the nearby smds (L3 D3 U3 on first picture ) . but the unit was working fine. when the 20.000uf cap arrived and i assembled the unit again the spark was very big and from that time the unit had the issue i will describe.

The issue:
screen works fine , menu works fine , buttons/adjustments in the menu works fine and stay in memory after restart , Vin shows correct on screen at 51 volts DC.
BUT : the screen shows 5,21 VDC all the time. if you try to adjust it , it works but the max voltage it will go is 9VDC and when i switch the ouput on and connect a load , the unit first overheats near the XL 7005 (the chip itself U3 or the coil nearby L3 on first picture) and then goes into shutdown and restart.

the voltages at the xl7005 are correct 5,21 vdc , the 8.4vdc dc2dc works ok , the input voltage is measured correct too , the output voltage is 5,21vdc measured all with precise multimeter.

my guess is that the 5,21vdc from the XL goes into output and it can supply only 0,4A so it overheats and shuts down , so shutting down the whole mcu/unit and then it recovers and restarts.
the datasheet of the xl says at GND pin that "great care must be taken to overcome sparks" so i guess that is the cause of the sparks i saw. but the XL seems to work fine.

none has a schematic/reverse engineered it , so the troubleshoot is impossible from me , i am not an electronic specialist . if you can guide me i think i can fix it. i do have 2 multimeters and an oscilloscope, and a hot air station/pistol to unsolder and an lcr/transistor tester to test the parts. but i dont know where to start and what to desolder.
(i think you need only the first picture , the error must be in thsi pcb , maybe the mosfets or the schottky diodes?)

some links and a pictures to help you
http://www.datasheetspdf.com/PDF/XL7005/839858/7
http://www.aosmd.com/res/data_sheets/AON6282.pdf
http://www.ti.com/lit/ds/symlink/lm5106.pdf

thank you for your time !

[KNSSoftware]

This must come close to not being economical to repair, when you  factor your own time in?  From a beginners point of view, it would make a good project from a learning point of view, but in that situation, you don't want the 'experience' taken away, and sense of achievement??

[BeetleJuice]

this unit seems to be very robust so maybe thats the reason none had ever problem with it. i google it and found nothing as an error for this unit.
add the 35$ and the easy construction and the tools i have and it seems cheap and easy to fix.
but i cant do it without help efficiently , yes i can start unsoldering all the parts and test them but this far beyond learning how this unit works

[arny]

I had one of these power supply modules with similar problem and after some investigation its turned out that LM5106 Half-Bridge mosfet driver has one defective output. you can check the two outputs using scope.

[BeetleJuice]

it was one of the three suspects ! i will check it.
can you please tell me which pins to check? i should check them with output on?

thank you very much!

[technogeeky]

I have... four ... five? of these units. No, one of them is a DP5005. So four.

One of my DPS5005 died due to back-charge due to a battery being on the output. I had accidentially swapped out the output cable that had a schottky diode to protect against this, with one that had no such diode.

The creator of the unit (who is a member of this forum, RD Tech) told me that there is a blown fuse somewhere. I haven't had the time or reason yet to find and fix it. So I am interested in what goes on here, but I can't help that much.

I will say, however, I did discover a method for disassembly of the double stacked unit. You just need two soldering irons, and two identical devices that have some outward springyness to them. By itself, this process isn't really useful at all (there are no components in the location I'm considering desoldering). However, this disassembly is necessary to desoldering the LCD screen connector -- which reveals a lot of important ICs.


First, flood and bridge the two sets of 10 pin holes on the back-most board. You want a big single solder ball that covers all of the 10 pins on either side. Then let this cool, and get a secondary soldering iron out. You insert the springs in between the two boards, right underneath the two 10-pin connectors. These need to have quite a bit of force on them.

Then, keeping the overall unit flat (be sure to protect the screen with some tape and/or foam or both), heat both bridged solder balls up as uniformly as you can, and let the springs work their magic. It will go a little bit more smoothly if you have a 2nd person nearby to hold the two boards clamped together (opposing the springs) until you're confident the pins and solder are both fully and evenly heated. That way you won't have to repeat this process, applying more and more spring extension each time.

I suspect there is some other method intended to be used to do this, but I don't know what it is or what equipment such a process would use.

[BeetleJuice]

thank you for the answer! maybe can we "tag" him (rd tech) to see my post?

I did charge some batteries but we are talking about 1 amp max at 4.2 volts and I can't understand how can a charged battery destroy the unit if the unit is on (i never left a battery on the unit with the unit closed). the unit had aa external diode to charge batteries, but it never worked, the batteries never get a charge with this diode no matter what polarity I choose.
but the unit was ok since i changed the big cap at the input of the unit and the spark got bigger. i think my unit had a problem since the first day , and thats the reason i saw the sparks. and when the sperk got bigger the unit had problem. thats my guess !

I think there is no need to disassemble the board, cause the problem is on the upper PCB. I do have however a hot air station and a desoldering gun and I can disassemble it.

my unit has a fuse F3 above D3, and that fuse is not in the initial photo I posted here. its strange cause when I measure the voltage, it has 2 different voltages across it (BUT its a fuse , so a single wire , so how can this be like a resistor , maybe its faulty) .

I post a photo with all voltages measured with the unit at 23 volts in and the output on.
today I will measure the signals (IN , HO , LO) with my oscilloscope.

thank you all for the help!

[RD Tech]

The story : this unit was purchased a year ago for 35$. i used a transformer 220->36 VAC 300VA , a 20A bridge rectifier and a 1000uf/63V cap 63V (ordered later an 20.000uf/63V).

the unit worked perfect BUT when i pluged the in/out connector i saw a spark between the connector and the nearby smds (L3 D3 U3 on first picture ) . but the unit was working fine. when the 20.000uf cap arrived and i assembled the unit again the spark was very big and from that time the unit had the issue i will describe.

The issue:
screen works fine , menu works fine , buttons/adjustments in the menu works fine and stay in memory after restart , Vin shows correct on screen at 51 volts DC.
BUT : the screen shows 5,21 VDC all the time. if you try to adjust it , it works but the max voltage it will go is 9VDC and when i switch the ouput on and connect a load , the unit first overheats near the XL 7005 (the chip itself U3 or the coil nearby L3 on first picture) and then goes into shutdown and restart.

the voltages at the xl7005 are correct 5,21 vdc , the 8.4vdc dc2dc works ok , the input voltage is measured correct too , the output voltage is 5,21vdc measured all with precise multimeter.

my guess is that the 5,21vdc from the XL goes into output and it can supply only 0,4A so it overheats and shuts down , so shutting down the whole mcu/unit and then it recovers and restarts.
the datasheet of the xl says at GND pin that "great care must be taken to overcome sparks" so i guess that is the cause of the sparks i saw. but the XL seems to work fine.

none has a schematic/reverse engineered it , so the troubleshoot is impossible from me , i am not an electronic specialist . if you can guide me i think i can fix it. i do have 2 multimeters and an oscilloscope, and a hot air station/pistol to unsolder and an lcr/transistor tester to test the parts. but i dont know where to start and what to desolder.
(i think you need only the first picture , the error must be in thsi pcb , maybe the mosfets or the schottky diodes?)

some links and a pictures to help you
http://www.datasheetspdf.com/PDF/XL7005/839858/7
http://www.aosmd.com/res/data_sheets/AON6282.pdf
http://www.ti.com/lit/ds/symlink/lm5106.pdf

thank you for your time !

Hi friend
thank you very much for your message
about the problem. I can't judge it 100%. but from voltage is alway 5.21V, max 9V,. It should be LM5106 problem.
but the unit first overheats near the XL 7005 (the chip itself U3 or the coil nearby L3 on first picture) and then goes into shutdown and restart. I don't why
please contact me by email 1749808860@qq.com
I can send you a new power part, you just replace it , it should be work , ok?
about the picture , it is old version, we have already change many things
and if you have any problem,. you can should contact us ..
we design this , we know the real problem and how to help you
please contact me by email

[RD Tech]

there is a new products was published again- DPS8005 buck power supply.
https://www.aliexpress.com/item/RD-DPS8005-programmable-constant-voltage-current-Step-down-power-supply-module-Voltmeter-Ammeter-buck-converter-80V/32849962454.html
there are 3 version, No communication version, USB communication version , USB+BT communication version.
Input voltage: 9-90V, output 0-80V/0-5.1A, Most function are same to other DPS series power supply.
But this time, there is new function for PC software, data download and data export.
You can download all file http://www.mediafire.com/folder/i1766ch7ec97a/DPS8005_file and there is a simple video : https://youtu.be/7SZcuMmvhtw
As the old rules, when new products was published, there is 4 days lower price.
Now no communication version 27.88 USD (normal price 33 USD)
USB communication version?30.43 USD (normal price 36USD)
USB+BT communication version: 32.98 USD(normal price 39 USD)
Only 4 days, please hurry… hope you like it
Lastly, there are 2 kinds power supply case, but they are not suitable for DPS8005, because the fan board MAX input voltage is 60v. but you can use the case, not use the fan board

RD official store https://rdtech.aliexpress.com/store/923042

[elektrinis]

Hello,

I am building a big lab PSU with 2x 50V 15A, with main PSU being a server 48V 2kW unit.

These are pretty noisy, so I would also like to add a 50V 5A linear regulator. 50V*5A = 250W of heat, obviously not the way to go.
So I am thinking of getting a 50V 5A unit and adding an LDO in series: buck takes care of efficient conversion and LDO takes care of noise.
There are some nice 5A LDOs with 250mV dropout. So ideally I would like to run the buck around +300-500mV above main output.

Anyone knows how the feedback is implemented on these 5A units? I see a half bridge driver chip, but no PWM controller. So I guess this is handled by STM?
The board also contains a 5V rail with XL7005A buck chip. Let's take it as example, took this schematic from datasheet. Can I modify this like shown here?


Hopefully this would keep the buck +0.4V above LDO.

Any thoughts about this working or not?