T/S Meanwell HRP-300-24

markz

100 TW
Joined
Jan 9, 2014
Messages
12,179
Location
Canada and the USA
Normal Operations are as follows
Upon connecting the XT90 's the charger will go full blast, stage A
In the last 5-15 minutes it will slow down and the sound reduce a hair, stage B
Then the final stage C it really slows down and the sound really reduces but you can still hear it.
All the sounds are constant and consistent at each stage.

Symptoms are as follows
1st full blast stage is fine.
If I move the charger it will still go fine.
If I leave it, and listen it seems that when its in its near or in the second stage it will sometimes go very low then back to normal with no movement of the charger. If it stays low, I move charger and it goes back to normal.

View attachment 1






Visual Inspection
Nothing sticks out, no black marks, no puffy caps, no missing pcb/case fasteners (could be a gnd)


Digital Multimeter
Continuity Check
Resistors Check
 
I have two other non-working MW HRP-300-24's stored away somewhere and I hope its just the soldered in fuse that went!

On the backside, its the very top and the 1st (has a test point next to it) and 2nd pins (direct above SMT at the circles 1oclock)
The smaller black box has continuity and is through hole. Its beside the yellow cap near the L input.
Text on the Fuse states F6.3A250VCQMSF right on the L Line from wall outlet cord.
Looks to be a TE5 or TE7 from my searches, probably a size format there.
Fuse on HRP-300-24 pcb board is 4.25mm wide x 8.15L x 7H measured with digital caliper.
TE5 spec sheet says, 8H x 8.5L x 4W mm
TE7 spec sheet says, 8.9x8.9x4.65

Again nothing out of the ordinary for the MW's that do work.
MW #2
View attachment 4

L Line.jpg



https://www.meanwell-web.com/en-gb/ac-dc-single-output-enclosed-power-supply-output-hrp--300--24
Manual - https://www.meanwell-web.com/content/files/pdfs/productPdfs/MW/Installation_Manual/Enclosed_EN.pdf
https://www.meanwell-web.com/content/files/pdfs/productPdfs/MW/HRP-300/HRP-300-spec.pdf

Meanwell Block Diagram.jpg





In "Report" link
"Testing Items"
https://www.meanwell-web.com/content/files/pdfs/productPdfs/MW/HRP-300/HRP-300-24-rpt.pdf




White powdery substance :lol:
Caked on, light fingernail dont do nothing.
by the top right circle, C1 to TVR 14471 yellow disc metal oxide
MW #1
View attachment 1



Found something interesting on my MW's that dont work. Doing two things at once here. One is fixing my erratic charging on my current operating MW's MW#1 and MW#2, but also I have two others that dont work.
There are two heat shrunk components, one longer then the other one, and they are side by side, next to two large caps. Well, there is no continuity between those components compared to the operating MW's. I will look into it further.

Toshiba K20J50D is a Field Effect Transistor N-Channel
https://www.glynshop.com/erp/owweb/Daten/Datenblaetter/Toshiba/TK20J50D.pdf

The Fuses are in operating order as there is continuity.




 
It can be pretty challenging to troubleshoot one of those depending on what happened.
I see you traced out the first stage. You should see continuity from the AC lines to the bridge rectifier. If you are brave, you can apply 120vac and measure the main input capacitor. It should be around 170vdc.

Beyond that, there are several common failure modes. Using the diode check function on a multimeter, you can measure the big power transistors and output diodes. The power devices are high fail parts. If the main switching transistor blows, it will frequently take out the PWM chip which drives it.

Another common fail is an electrolytic capacitor that has gone open. A meter with a cap check function is handy for this.
 
Thanks for the tips fecther
I didn't know what else to do other then continuity and resistance measurements.
It all looked quite simple from the top of the pcb, however the bottom is very populated.
Going methodically step by step through each block/stage can be quite daunting.

I came to an Ah Ha moment, when the charger is in use everything heats up as normal, expanding and contracting joints.
Another was the charging wiring ends at the screw down output looks to be tarnished but there feels to be bulk of strands where the solder squished. A new wiring harness, new connectors is todays goal. Just gotta pound out a few things in the light snow today.


120V input from L Line.jpg



R8 is open and needs to be replaced on my non working MW's.
big diode, bigger diode top left.jpg




Charging connector.jpg



Light Grey almost hidden "CAP"
Needs white lettering.
Digital Multi Meter.jpg




fechter said:
It can be pretty challenging to troubleshoot one of those depending on what happened.
I see you traced out the first stage. You should see continuity from the AC lines to the bridge rectifier. If you are brave, you can apply 120vac and measure the main input capacitor. It should be around 170vdc.

Beyond that, there are several common failure modes. Using the diode check function on a multimeter, you can measure the big power transistors and output diodes. The power devices are high fail parts. If the main switching transistor blows, it will frequently take out the PWM chip which drives it.

Another common fail is an electrolytic capacitor that has gone open. A meter with a cap check function is handy for this.
 
When I have the meter on diode/continuity, its reading which side is Anode (positive) and which side is Cathode (negative) but I also get a reading, non powered up of course.
D5 is 411
Hyper Fast Diode (A K) is reading 411 as well.
What do those numbers mean?

Basic Video says 500-900 reading is a good diode.
https://youtu.be/qSDRsz5-t7I?t=80

Measuring Capacitors
I do need shorter leads. Dont have them yet.
Move red probe to Temp/mA
Unpowered I measured nothing on all ranges, they are open. Big brown caps, all MW's.
Time to buy shorted leads.
Did same with red lead on 20A, nothing.
 
A diode check reading of 411 indicates 0.411v across the probes. This would be a good reading for a Schottky diode. It's good to swap the probes around and verify the reading is higher in the other direction. Sometimes other components in the circuit will affect the readings and you may need to lift a leg in some cases to isolate it. Diodes typically fail shorted unless they get blown apart.

The open R8 likely indicates the FET is blown.

A good FET will measure like a diode between the source and drain and like an open circuit on the gate to the other legs. Failed FETs generally short between legs.

Any part that gets hot during use expands. Repeated expansion and contraction can cause fatigue failures in the solder joints or even the legs. With close examination, fractures in the solder are often visible. When in doubt, resolder the connections.
 
I see

When I was measuring resistance of R8, what I was really doing was measuring the resistance of S and G through the Field Effect Transistor N-Channel.

I got it all back together, made some safety add-ons for the series connection V- to V+
Later I will be using a fatter insulated wire, 15A runs through that.

As I was putting it back together I was thinking I should upgrade the conductor size of the 120Vac input cable.

I will see what I can do for a video or audio of the charging process, I never done it before to add to the internet.
Maybe I can do it with a Google drive account.


20.20V + 21.80V = 42.10V and each one is the lowest it goes.

Back together charger.jpg
 
Video
Hey it's working! Dont mind my noises 8)
https://drive.google.com/open?id=1KA5464IBHwej_T10LDJk78DNdKHHu46g

1st audible change is at 25:00
Rotated around and nothing
Very slight oddities in sound and pitch at 27:00
Very Noticeable and not normal at all at 28:00
Plug into second battery while I go get my DMM at 30:40
Measured 1st battery at 42.20V

Now you are going to ask yourself why I slightly banged the charger, I know its not good but just wanted to see because sometimes it would go again then sputter.

Constant Current up and into second stage.
Second stage lasts for 3-5 minutes, pretty much full on so instead of 15A perhaps its probably 11A, will measure later.
CC is fine up until it reaches set voltage & equilibrium. Constant Voltage stage is where the charger is having issues.

One thing I would do, but I can't, is set the voltage to 39V.
Once I get one of my non-operational MW's working, then I can start isolating which MW is acting odd. In the meantime, finding a FET (N) is the name of the game.
 
(S) Source (D) Drain (G) Gate
Red probe on S + Black on D = 502
Red probe on S + Black on G = 203
Every other combo is open.

fechter said:
The open R8 likely indicates the FET is blown.

A good FET will measure like a diode between the source and drain and like an open circuit on the gate to the other legs. Failed FETs generally short between legs.
 
It was neat to troubleshoot for a bit, get the juices flowing again but it is daunting. Just wanted to see if it was something simple, atleast now I know there is a soldered in fuse, and can do a few simple checks.

I will buy some SMUN MW clones

Can you leave them unattended like you can with the genuine MW's. Genuine units will just match the voltage and in time the green led turns off.


Hillhater said:
What is your time worth ?
These things are ~~$25 a piece or so.
Why f' about fault finding...just order a spare or two.
PS. My 24/300 clone will run down to 9v (hums a bit at that voltage !),..and up to 30v.
 
I never charge totally unattended....but i dont "hover" over it either !
On my clone, the it definitely does a CC/CV charge profile, but green light stays on at full charge.
However, the current tails off to something unmeasurable (0.01 A?)
 
Hillhater said:
I never charge totally unattended....but i dont "hover" over it either !
On my clone, the it definitely does a CC/CV charge profile, but green light stays on at full charge.
However, the current tails off to something unmeasurable (0.01 A?)

Have you ever connect two in series?

Asking cuz I need to build an outdoor PSU LED driver for battery heater.

My 24/300 clone will run down to 9v (hums a bit at that voltage !),..and up to 30v.
SMUN arlen68 states 24V is 21.6-26.4V so I think even with one of my meanwells its a few volts higher on the top end then all the rest. Perhaps just lucky with your 30V?
 
Well a few things. Bike fell over twice in the wind, messed up a +V fastener but I got it working.
Decided I will just solder them solid, replace the top cover, and install XT90 Anti-Spark connectors, with longer leads.

The XT90AS female connector has the green on it, there's nothing inside the male that comes in the package right.
Batteries have normal XT90 male connectors.
In time, I will find out.

Next up I just need to slap my two batteries into one, makes for a healthy discharge, instead of draining one then hooking up the other. Shallow is best. First I want to see how far I get with both at a gentle speed, my last measurement was 22km with the big pack, and 15km with the smaller pack. Temp tomorrow is +8C, route will be the same, flat as can be.



Had to do it!.jpg




View attachment 2




View attachment 1




Ready to go.jpg



Much longer battery leads, double the length from last. A few juice up spots require longer cables.


I have one or two problems
1) Charger is not charging completely, but I measured and its pretty much full by the time the charger acts weird.
2) Battery packs are severely reduced in capacity, this is likely the case.

Analysis of my Main Battery Pack
https://endless-sphere.com/forums/viewtopic.php?f=14&t=96683#p1416581
 
Day is gone, cancelled meetings.
No spare backup :(
Time to get on this!

Heard a pop, while charging my new battery 10S10P
Battery is now at 40V, was charging from 36.8V for 15 minutes before I heard a pop.
First charging cycle on the soldered charger, first charge on the new 10S10P configuration.

Hope its just a fuse. Hope its not too hard to find the bad component.

Upon a chance encounter while getting ready to de-solder ac input, I found a stray strand from AC ground to case.
The others ones dont have continuity between ac ground and case.
I know the pcb ground is connected to the case.
Case ground, is input ac ground.
Not good, not good at all.

This is what I know, the ac input gets to 4.5 - 5A.

At the time of charging the fan was still blowing hard, typical sounds of full-on stage.
Then POP! no smoke, cant remember if there was a smell.

This is the best pic I could get.


culprit.jpg
 
Thats right!

Tunes-------------------------------------------------------------------------------------------> Check
Cherry Pepsi-----------------------------------------------------------------------------------> Check
Windex'd Desk---------------------------------------------------------------------------------> Check
Strain Relief cheap ass Princess Auto soldering iron------------------------------------> Check
Digital Multi-Meter with no rear stand in a half decent angle for viewing----------> Check
Do my arrows all line up?--------------------------------------------------------------------> Check





 
Five Meanwell HRP 300 24
One Working
Four Not Working


This is what I was thinking.

There are 5 pcb mounts to case.
Electrical/Electronic design of mw, I think they should all be connected, meaning leave pcb in for now for a power up t/s.

Meantime
Check for obvious signs.
Fuses

Back and forth, good to bad, bad to good.

Top cover off
Fan out - Note: In factory layout the wire routing goes under the wire for the coil (Black heat shrunk with 2 wires coming out)
 
I labeled the mw's from 1 to 5.
mw1 is the good one, which was on the bottom of my setup.
mw2 is the top one, always took the most abuse. Is also the one with the rogue wire strand, causing mw2 to pop!
All are labeled on the big cap, next to where the fan goes.








Also
Bigger drink cup for pepsi and club soda mix.


GAME ON!!!!
 
A bunch of screws come out from the sides.
5 bolts for the pcb to case.


fix1.jpg


New snag, mw1 is non operational :(
Alligator clip slipped off ac input.
[strike]Fuse[/strike] Diode is broken Big black surfacemount, in picture right edge, going down from top most Right, 3 through holes, then 2 through holes then fuse.

Orientation: When case is: input/outputs front and fan rear.
One metal plate top left.
One square piece of rubber with white paste, below that.
 
[strike]Fuse[/strike] DIODE
Big Black Diode is LT 1228 S3 J

mw1 and mw5
- As seen in pictures:
- Red probe down, black probe up -------- is good
- Red probe UP, black probe DOWN ------- is no good

All other mw's have a reading both ways ~450 to 500 on continuity setting.




 
Alrighty next up
"Green Disc" going into Big Blue Relay
Not a classic resistor as I am used to.
The pcb has a resistor squiggly. As you can see in the picture below.
A quick search reveals: A thermistor is a temperature sensitive resistor.

No markings on either side.

----------------------------

OK so I'm going to check that little black box fuse coming straight off of "L"
Markings say on PCB say
F6.3A
/250V
FS1

Component says
F6.3A 250V CQMSF

----------------------------

Results of "that little black box" fuse is as follows

mw1 - Good!
mw2 - BAD FUSE
mw3 - BAD FUSE
mw4 - BAD FUSE
mw5 - Good!

*Yes, I double checked, 8 times over!

----------------------------




 
Thinking all this tedious work can help others out.
BTW I am looking for a good, free, circuit design software, dont got time right now, I did a cursory search.
Having a hard time getting ngspice going.
This one is too basic http://www.partsim.com/simulator#


Newest Schematic




schematic.jpg




MW's.jpg




Fuse1.jpg




 
Through Hole Fuse at "L" AC Input

Vishay Semiconductors RST 6.3 and RST 6.3 AMMO
RST 6.3 is Bulk are loose resistors.
Ammo is taped on both ends and fan folded in a box.
https://www.mouser.ca/Circuit-Protection/Fuses/_/N-ba85y?P=1yyc4ikZ1yyc4etZ1yy2lknZ1z0x8dp




D5 Diode
LT1228 S3J
S3J is the series

6.75mm x 6mm component on pcb

A search of S3J came up with this
http://www.vishay.com/docs/88713/s3a.pdf
"Surface Mount Glass Passivated Rectifier"
Mouser website Vishay Semiconductors S3JHE3_A/H and A/I
Cant find the difference maybe how its on the reel
https://www.mouser.ca/Semiconductors/Discrete-Semiconductors/Diodes-Rectifiers/Rectifiers/S3J-Series/_/N-ax1mb?P=1yxk9wo&Keyword=surface+Mount+Glass+Passivated+Rectifier&Ns=Brand%7c0&FS=True


----------------------------

It looks to me as though the pcb board itself is 2 layers.
I can see through it when I hold pcb up to the light.
Top layer
Bottom Layer
The pcb itself viewed at an angle, looks to be a line through the middle

pcb layers.jpg

----------------------------

Plan A is to buy the fuse listed below locally, its not 6.3A but I think 5A will do. Hopefully this will fix 3 of my mw's.

https://www.be-electronics.com/product_p/sr5f-5a-bk.htm
Buss Fast Acting Fuse Radial 5A 250Vac PCB Thru-Hole style 5.08mm
# SR5F-5A-BK $1.58 (buy 5)

Better to have a fuse then not.
Plan B is to solder in a wire on the through hole of the fuse
 
Back at it
Bought 10 fuses, not the same 6.3A vs 5A hopefully its not an issue installing a lesser amp rated fuse.

The plan, used a needle to open up the holes. Pliers, flat heat, robertson, took too long but got it done.

fuse replacement.jpg

Not my best soldering job

View attachment 2




soso1.jpg




 
There is a reason the went with a rectangle square. Mine are round, so put a piece of paper in there just to be on the safe side.
This soldering job aint my best work. Cheap Princess Auto/Harbor Freight 40W iron, some thick ass solder I got because I was fed up with thin ass solder.



Fasteners
Mini bolts with washers (x5) for the pcb to case mount.

Longest bolt goes to the temprature probe with 2 white wires, heat shrink on both so it doesnt short to the case gnd. wire routing goes between the cap and the coil, and temp probe connector facing face, wire doubles back. Makes sense, dont want it to be by the fan, and if the probe connector is the other way it would be touching the coil. ---> Do not forget about the plate with 3 holes it. That plate, goes between the case and the components. See pictures. Heat transfer from components to case. And yeah my cases can get hot to the touch sometimes. Two smallest bolts secure the plate, longest bolt secures the probe.


2nd longest bolt and bar goes to secure the other two fets by the fan. Both heat shrinked. Bump on bar goes out, rounded off side faces component.

Square heat shrink goes to to by the blue component, behind the rectifier, 2nd longest bolt and bar, bump out, rounded side in.

Fan has two big chunky bolts, wire routing is top Potentiometer side. Red wire to screw terminals. Wire routing I believe went under white wire. Label out.



done and continuity yaaaaay.jpg




goes in behind for extra heat dissipation I think.jpg




lots of white stuff goes to the case side.jpg




View attachment 1













Urghhh!
Somethings not right either 5A is too low because the charger requires more then 5A on plugin with no battery
and need to match to the 6.3A. Slight sound of a pop, recheck 3 times and fuse is no good.
or something else is up, another component close to the ac in.

Fuse is Conquer Electronics Co. Ltd.
Thats probably what the CQ stands for labeled on the fuse itself.
MSF6.30 CONQUER ELECTRONICS CO. LTD MSF 6.3A 250V Q/B BULK
List http://www.anglia-live.com/products/Circuit-Protection/Fuses-Breakers-Holders/Standard-Fuses#Brand%2c-30|Current+Rating%2c3011|Mounting+Type%2c3470|Voltage+Rating%2c13203|resultsperpage,63||pf,Brand%2c-30

Exactly the same http://www.anglia-live.com/products/circuit-protection/fuses-breakers-holders/standard-fuses/1156877001_MSF-63A-250V-Q-B-BUlK
^^PDF
http://www.anglia-live.com/netalogue/pdfs/cqr/datasheets/MSF%20Data.pdf

CONQUER
Conquer Electronics Co., Ltd.
Micro Fuse
Type MSF
Fast Acting Radial Lead Micro Fuse Series


fuse pdf.jpg
 
Back
Top