power supply ?
Hi all,
Next project - after I put the mixer back together...
Powertech Plus, MB-3623, 6volt, 12volt and 24volt.
No display at all. Stone cold dead. When the pos and negs touch no sparks, nothing.
Who wants a guess as to what I'll find inside this and is at fault before I open it up, I'm doubting a fuse, but you never know?
Look Here -> |
power supply ?
Skepticist (08-03-19)
Yes the control circuit power supply and how it's derived is the first thing to look at here.
Also the potential hazards are unknown so treat it with maximum respect until there's enough understanding to approach it safely.
hinekadon (08-03-19)
Crafty buggers have got a security screw in one of the six corners
About 50mm down and the access hole is only 5mm across. I have a hex set that would fit into the screw, but the diameter is 8mm so too thick to go in the hole..
I've tried to put a groove/slot in the middle in two small flat head screwdrivers, but the hacksaw doesn't touch the metal
Any tips on how to get the thing out so I don't break anything to get to it?
aha been there i put an old part of an radio areial down the hole and gave it a bash , the end scrunched up to the shape of the screw and loosened it as well so it came undone easily don
VroomVroom (12-03-19)
They really really don't want anyone poking around in there it seems and no matter how many security bits you have you'll always be lacking the one you need.
That one doesn't look too hard to craft your own from a mild steel rod though with a dremel type tool, carbide disc and a blowtorch for tempering if needed.
You could also cut a groove in the shank end of a drill that fits in the hole - that end of a HSS drill is not as hardened as the business end (can easily turn down the diameter of large drills to fit them in a smaller chuck)
With patience, determination and a good hacksaw blade, results may be forthcoming.
Or am I seeing things because the picture isn't clear as to whether it's slots or pegs - in fact slots make more sense now so the tool needs 2 rectangular pegs? Still a homebrew possibility if so IE could butcher a screwdriver.
I like challenges
Last edited by Skepticist; 08-03-19 at 12:36 PM.
hinekadon (08-03-19)
Why oh why do I get myself involved in these things...
And the mixer looked easy!
nomeat: Sit back. I might actually kill myself on this one
So all apart. With the control panel detached, I decided to measure where 240v was running around.
First unusual question: Why am I reading 30vac when I have ONLY the pos probe on A and when I touch the neg probe with my finger it reads 180vac and I think I can feel a very slight tingle.. I know I'm gonna die on this one
PCB seems to be in 3 power sections:
So: 240vac is virtually everywhere. The large solders most are all reading 240vac, even down to the B- on the lower section.
Used drill bit, thanks Skepticist, a work of art if I don't say myself....
Nice bit of toolmaking there
Please work on it off-line until some sense can be made of the circuit.
Seems to be a lot of accumulated dust so clean that all up first - maybe compressed air will do the job
It's a Jaycar unit - MB3623 is its catalog number - definitely search online for any info related to problems
From a cursory look at the pics it appears to have 2 switchmode supplies IE the large one for the charger output and a smaller supply presumably for a 'standby' type DC output for the control panel etc.
There are optocouplers so the secondary should be isolated from the primary side.
There's a fuse with black heatshrink over it so certainly check that for continuity .
And definitely check that large Mosfet clamped on the heatsink close to the larger transformer - the gate terminal soldering looks a bit odd in the pic (black hole?)
The circuit seems well laid out and the markings are clear
Check out how the mains Earth is connected - not a good reference for taking voltage readings but don't go livening this up before checking the fuse, the NTC thermistor, bridge rectifier and Mosfet at least, the first thing to look for in a live test is the 340V DC out of the rectifier - ULTRA carefully! and not using ground as reference (need to check across the HV capacitor) and only after all those HV side components are tested and proven OK
I'd also be using a series lamp setup to limit any fault currents
DISCLAIMER: I'm just verbalising my thoughts on how I'd approach this, not suggesting that anyone else should follow my ramblings
Last edited by Skepticist; 08-03-19 at 09:38 PM.
Look what I found;
It certainly had gone and my heart jumped with joy at how easy repair this was going to be
Until after I had replaced the 4A, connected the front control panel, plugged in and - BAMM!!!!
R11 blew up!
It's placement on the underside...
Looking around the board, it looks like water damage. Corrosion all over. Obviously a short somewhere...
I'm thinking it's not going to be worth the time to do this one, the testing bit by bit (considering the mixer). I have to admit I don't have the skills to do more in depth testing and the components are all fairly tightly packed and with HV running around.
Probably lucky you didn't get burnt in that one.
Fuses seldom fail spontaneously and that's why I was suggesting checking up to the mosfet prior to any attempt at energising the circuit and, even then, using a series lamp to defend against any blow-ups. That R11 looks to be 0.22 ohm so likely to be the overcurrent sensing for the switching controller and that's what would happen to it if the mosfet was a short cct and the fuse took too long to blow.
GT250 (09-03-19)
Is this the Mosfet? Taken from next to R11. Right leg was attached to R11.
The Ω readings are (with neg on center leg), Left leg 74Ω and Right leg 32Ω.
The R11 is definitely gone so it's 0.22Ω? Red Red Silver Gold so 2 2 x0.1 = 0.22Ω Is this a 0.5watt one or a 1W?
Last edited by GT250; 09-03-19 at 12:03 PM.
Yes 0.22 ohm 1W and that's the Mosfet (900V 8A 160W) datasheet:
Mosfet appears to be toast and there are lots of suitable replacements for EG any mosfet rated 600V or higher, Id 8A or higher (if an exact replacement is hard to get)
You could be lucky with this charger if the fault is limited to the primary side (Fet & fuse & cleanup) & resistor
Is the Fet's drain insulated from the heatsink? IE if it has exposed metal on its back, was an insulating washer used or was it mounted straight onto the metal heatsink?
Last edited by Skepticist; 09-03-19 at 01:46 PM.
I have rarely, no I think never, seen a power MOSFET fry in an SMPS or inverter without taking the driver circuit with it. The transistor might be replaceable but not the driver IC.
A fuse that looked like that is also never a faulty fuse but always catastrophic failure elsewhere.
And by replacing the fuse and powering it up without testing main components like FET, diodes and large filter caps has almost certainly nuked the driver circuit.
This is certainly not the type of projects worth taking any risks.
I am sure you will find better DIY projects in the near future.
Last edited by Uncle Fester; 09-03-19 at 01:52 PM.
Update: A deletion of features that work well and ain't broke but are deemed outdated in order to add things that are up to date and broken.
Compatibility: A word soon to be deleted from our dictionaries as it is outdated.
Humans: Entities that are not only outdated but broken... AI-self-learning-update-error...terminate...terminate...
betcha ten bucks it isnt .
Put your meter on beeper and check the FET legs 1-2, 2-3 and 1-3. (DO NOT connect Mains Power)
It will beep if the FET has a short and is then no good.
I wouldn't give up on this so quickly as it's a simple flyback design and I've encountered failed switching transistors without damaged control circuits quite a lot. It depends on what caused the failure and how the control circuit derives its operating power supply. Overload/overheat failure tends to just take out the switcher but overvoltage (EG lightning) can take out a lot more.
the first thing to test is if its getting to frequency cos the transformer is a short circuit at dc and a load at ????300hz???? to achieve this the output from the fet has to be set upon a resister not the transformer with the same impedance as the tranny at its freq if the oscillator starts ok then check the load from the rectifiers for shorts you have two to check each time one high current and one controll supply looking at the fuse it appears to be a short by the way its gone , so i would start at the high current circuit good luck
That might have been possible before he replaced the fuse. The built in zener protection in the FET could have saved it but after he fired it up with the new fuse the FET just fused together and blew the resistor.
Didn't you see his ohm readings?
Gate would have been literally exposed to 400V for a split second before the resistor blew which goes straight to that little controller board that looks like it has an 8 pin IC. They can only handle a few 10's of volts.
I can sometimes find those chips on the net but minimum order is usually 1,000 or 10,000 pcs.
Update: A deletion of features that work well and ain't broke but are deemed outdated in order to add things that are up to date and broken.
Compatibility: A word soon to be deleted from our dictionaries as it is outdated.
Humans: Entities that are not only outdated but broken... AI-self-learning-update-error...terminate...terminate...
Ok.
Went to Jaycar and they stock neither the R11 or the Fet!
Nomeat: I get it
Loopyloo: Yes, it beeps on all. As tested out of the board.
Hinekadon: I agree with what you said - just a bit above my IQ level and the "so I would start at the high current circuit...", Err no!
As I would have to order in R11 and the Fet + Postage and time, I'm not going to continue, and as nomeat has said other things in the pipeline could be at fault.
If anyone would like to have me post it to them, I'm ok with and I'll even pay for the postage as long as you don't live in some remote location and I hope you get working - just a shame to chuck it if it's a simple repair for someone more qualified than me.
The price in the Jaycar catalogue is showing $159.
Many thanks to all who helped.
Bookmarks