HANNY Switching Power Supplies

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
User avatar
jollyrgr
Posts: 1289
Joined: Thu Jan 03, 2002 1:01 am
Location: Northern Illinois
Contact:

HANNY Switching Power Supplies

Post by jollyrgr »

Anyone here read Chinese?

I am trying repair a bad power supply in a mini refrigerator. The power supply is bad; there is no output. This is a SWITCHING power supply that uses 120VAC input and converts it to several amps of DC output at about 12VDC. It is a HANNY (brand) FX-101 (model/part number). I've visited the site (http://www.hanny.com.cn ) but cannot find either the supply or a datasheet. At least on on the "broken English" part of the site. I DO know they make the supply as their name is clearly on the supply. Other info on the board is PCB50427E1.

The supply takes in 120VAC converts it to 320VDC (at least across the two switching transistor's heat sinks) and is supposed to output several different 12VDC in the final output. On the "high" voltage side everything appears working. On the DC side there are two ICs, several LEDs etc.

There is no output on the DC side. There IS an input for controlling the power supply and the controls appear to work but the control input does not appear to matter. There are two ICs on the main board, a TL494CN and a LM358. Neither have Vcc voltages. It appears as if there is no DC voltage on the "low" side what so ever. Since this is a non standard supply I need to find a data sheet for this board. This company makes OEM boards for a number of devices and appear to have data sheets; but I cannot read Chinese. I've done a number of YAGOOGLE searches and keep finding the company but not the board itself. A number as simple as FX-101 is not much help as it does not seem to make any headway into the search. I've translated the page using online translations; not pretty.

Can anyone help?
No trees were harmed in the creation of this message. But billions of electrons, photons, and electromagnetic waves were terribly inconvenienced!
User avatar
Chris Smith
Posts: 4325
Joined: Tue Dec 04, 2001 1:01 am
Location: Bieber Ca.

Post by Chris Smith »

I would start with a good DC car battery and ammeter, if that’s what your looking for.

Measure the DC and go from there.

There are a million ways to get DC after that, including a computer power supply if its within range, or the old fashion bridge rectifier / transformer which works but uses a tiny amount more power and more copper parts?

[Ohh well? ]
User avatar
Bob Scott
Posts: 1192
Joined: Wed Nov 20, 2002 1:01 am
Location: Vancouver, BC
Contact:

Post by Bob Scott »

If Hanny were in an english speaking country, I would just phone and ask to speak to an applications engineer and ask for the datasheet. I've done this before with another switching supply manufacturer. Web sites tend to be sales oriented and you'll probably only find info on the merchandise that they are currently flogging, not on stuff that's a few years old.

Since they are in China, I suggest email. Here is the contact list from the www.hanny.com.cn website.

E-mail:[email protected]

http://www.hanny.com.cn

http://www.head-ele.com

http://www.head-supply.com

Regards, Bob :cool:
User avatar
Edd
Posts: 885
Joined: Sun Jan 27, 2002 1:01 am
Location: Dallas Tx
Contact:

Post by Edd »

.
.
der Jolly Rojerewski:
Hows about a further analysis and the looking at the primary/hot side of that power supply and confirm that '494 might be its tandem driver for that pair of power devices connected to its power transformer.
With consideration for its drive outputs to terminate into the power devices from its pins 8 and 11. The other consideration is to make a foil trace path sleuthing from its pin 12 and to see if it terminates into a combinational of a line connected series droppping resistor to then feed into a rectifier for raw DC power with its then being filtered by an lectrolytic and passing thru yet another current limiting resistor to then be shunted to ground via an ~ 1 watt zener diode with a final ~.1ufd ceramic hf bypass, paralelled with yet another decoupling electrolytic filter. That resultant DC then being your '494's Vcc power sourcing. ( No I.C. Vcc power....shorted zener diode ? )

The other thought on the design would be its utilization of a brute force power oscillator on the primary/hot circuitry with the secondary ...past the transformer...cold circuitry to have utilized an optical coupler tying back into the primary loop ...isolatively....to initiate increased oscillator power enhancement in any increased current demand situations.

73's de Edd
[email protected] ..........(Interstellar ~~~~Warp~~~Speed)
[email protected]........(Firewalled*Spam*Cookies*Crumbs)


Hapiness is finding out what you like doing best and getting someone to pay you for doing it.

Image
User avatar
haklesup
Posts: 3136
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Post by haklesup »

Whenever a DC supply goes down, especially a switcher or anything current limited, you need to disconnect the load before pronouncing the PS dead. Naturally if the load is shorted the PS will either current limit to a low voltage (obeying ohms law) or self destruct. ALmost all switchers will have built in overcurrent protection.

Since you are talking about high current DC to supply a fridge, it must be using a peltier device. In which case you can check to see that it works like a giant diode. THis can also be tested with a sufficient battery but use a 30A fuse or circuit breaker in case it is shorted or you may end up with a lot of smoke as all the wire's insulation goes poof (not to mention burned fingers) Battreries are inherently not overcurrent protected

Does the fridge use 320VDC or are you just referring to a node inside the supply that does not go to the load?
User avatar
jollyrgr
Posts: 1289
Joined: Thu Jan 03, 2002 1:01 am
Location: Northern Illinois
Contact:

Post by jollyrgr »

Thanks for the replies. This week at work has been terrible so I've not had time to get back and check this thing yet.

The refrigerator is indeed a Peltier device. All of the DC side of the supply EXCEPT the Peltier can be unplugged and I've done this. I don't hear a "crowbar" effect like when a supply is shorted but that could simply be this supply.

The 320V is across two heat sinks for two transistors on the "high" voltage side of the power supply. This does not feed the fridge but is on the AC part of the circuit. My guess is this DC voltage is chopped into high frequency and then fed to the small transformer.

The circuit board has a distinct line on the trace side of the board showing the HIGH and LOW voltage sections. Unlike a TV where there is plenty of space (relatively speaking) this supply is kind of tight. The AC line is not isolated and there are two heat sinks with a 320VDC between them which could easily be grabbed or be touched accidentally. (These two heat sinks are less than a 1/4" apart from one another.) There is no way I was going to try and troubleshoot this nasty device tired. Too easy to make a mistake. Considering how bad this week has been, I would have made a mistake.

A quick check one morning, though, made it appear as if the switching transistor on the LOW voltage side is shorted. That being the case I think I'll take Chris' advice and just find a different supply and mount it external. The transistor is coated with heat sink grease and some sort of potting so it cannot be read. (Strange as everything else is fairly easy to read; just not this one transistor.)

I had pulled all the fans off the board as they had conectors. The Peltier did not; it is hard wired. I'll unsolder that and see if the supply comes up. But correct me if I'm wrong; don't switchers require SOME load to function?

More info to follow....
No trees were harmed in the creation of this message. But billions of electrons, photons, and electromagnetic waves were terribly inconvenienced!
User avatar
jollyrgr
Posts: 1289
Joined: Thu Jan 03, 2002 1:01 am
Location: Northern Illinois
Contact:

Post by jollyrgr »

More info....

Brief reverse engineering and testing reveal:

AC line goes through a fuse and a parallel wound choke. Some sort of HASH filter. (Two coils wound on the same form. AC line passes through each coil. NOT like a transformer but like two coils; one in series with each side of the AC line.)

The "choked" AC is then fed to a bridge rectifier arrangement. But there is a jumper such that it converts this into what appears to be a voltage doubler consisting of two diodes and two electrolytic capacitors in series. This is where I get the 320V across the heat sinks as mentioned before. Removing the jumper causes the voltage to drop to 160 volts. Checking the Internet shows this is a common way to make a dual voltage (110VAC 220VAC) switcher.

It appears that there are TWO "low voltage" switchers fed from this DC supply. Each having its own set of oscillator transistor pairs, transformers, and double diode rectifier. One power supply appears to be a low current, lower voltage dedicated to supplying the Vcc power to the TL494 for and LM358 ICs. This supply consists of two transistors in TO-92 cases, a smaller (of two) transformers, and two discrete diodes.

The higher voltage (12VDC), higher current (six to ten amps?) consists of a double diode with common cathode in one package. This is the supply that powers the fans and Peltier. The rectifier diodes are not shorted and appear fine. There does not appear to be any voltage on either power supply even with all driven devices disconnected. There does not appear to be any voltage, DC or AC at the transformer inputs (when measured with a DMM). So I've isolated the failure to somewhere after the "high voltage DC" supply and before the inputs to the transformers. I'll assume there should be some sort of high frequency oscillations from the transistors. Considering there are two switching supplies, each with their own oscillator transistors, transformers, rectifier diodes, etc. I don't think there is a failure in the transistors. (Plus the Ohm meter check shows the HIGH LOW readings as expected with a good transistor.)

I'm not real familiar with switchers and the tutorials I've read have thus far have not revealed a supply similar to this one. So far I've not pulled out the O'scope as this is is a supply without an isolation transformer and I don't have one to use. (Worse case is to use a UPS unplugged from the AC line to supply power while testing with the scope.) Thus far I have been using the negative of the caps in the voltage doubler as a "ground". Good or bad choice? (Remember; I'm working without a schematic except for what I've mentally reversed engineered.)

Short of checking every component with a meter one by one, is there any suggestions of what could have failed between the HIGH DC and the input to the switchers that would be common between two switchers?


Thanks for any replies. This one appears to be kicking my butt but I don't intend on letting it beat me. ;-)
No trees were harmed in the creation of this message. But billions of electrons, photons, and electromagnetic waves were terribly inconvenienced!
User avatar
Edd
Posts: 885
Joined: Sun Jan 27, 2002 1:01 am
Location: Dallas Tx
Contact:

Post by Edd »

.

Well, your first two paragraphs of circuit design analysis are right in line with a typical 'puter SMPS. Its configuration capability of the power rectifiers / raw DC filtering can accomodate either 120 or 220 VAC operational capabilities. Plus hash filters, line fuse and line Mov'(s).

The smaller circuit being the low power supply circuit for its ancillary PS support components and the second larger supply being the main power output with its single voltage / high current windings output .Whereas, the computer power supply would have its myriad of multiple windings for the production of +3.3V +5V +12V -12V outs.

Inquiring again, did you find any interconnect thru the cold power supply side and the hot side via either optical or inductive coupling means ? Can you trace the bases of the power xstrs to see if they intercouple in a power oscillator configutaration or are dependant on separate / isolated drive power into them?

Also, are you the initial owner of this system, such that you are aware of its age / performance history, or is it an exotica curbside / Dempster Dumpster diver find ?

Since your description of its components conditon doesn't sound like an experienced hardcrash of the system, I would suspect gradual deterioration of critical electrolytics within the system.
With that particularly true of either developed voltage DC filters on the secondary cold side of the suppply and / or any units used in coupling application. Electrolytics really take a HAMMERING in SMPsupplys.
There is a gradual higher heating level and electrolyte temp / pressure rise and some seeping out and then progressively further incremental compounding of the heating and the eventual production of a high ESR unit that eventually appears transparent in circuitry coupling and bypass
functioning. Thus a system failure, yet without any massive power component damage being inflicted.

Sooooooo, does you has either an Eee--ess--Ruh meter OR a function generator/or/audio oscillator/or pulse generator of 50 0r 100 Khz capability ? With that and your old- silly- scope, 2 toothpicks, some Scotch tape and 2 clarinet reeds, I can show you how to make a F16 screaming tree top approach, totally under the radar.

Awaiting further feedback.

73's de Edd
[email protected] ..........(Interstellar ~~~~Warp~~~Speed)
[email protected]........(Firewalled*Spam*Cookies*Crumbs)
:grin: :razz: :grin:

If you keep your feet firmly planted on the ground, you'll definitely have trouble putting on your pants.

.
User avatar
jollyrgr
Posts: 1289
Joined: Thu Jan 03, 2002 1:01 am
Location: Northern Illinois
Contact:

Post by jollyrgr »

Edd wrote:.

Inquiring again, did you find any interconnect thru the cold power supply side and the hot side via either optical or inductive coupling means ? Can you trace the bases of the power xstrs to see if they intercouple in a power oscillator configutaration or are dependant on separate / isolated drive power into them?

Also, are you the initial owner of this system, such that you are aware of its age / performance history, or is it an exotica curbside / Dempster Dumpster diver find ?

Since your description of its components conditon doesn't sound like an experienced hardcrash of the system, I would suspect gradual deterioration of critical electrolytics within the system.
With that particularly true of either developed voltage DC filters on the secondary cold side of the suppply and / or any units used in coupling application. Electrolytics really take a HAMMERING in SMPsupplys.
There is a gradual higher heating level and electrolyte temp / pressure rise and some seeping out and then progressively further incremental compounding of the heating and the eventual production of a high ESR unit that eventually appears transparent in circuitry coupling and bypass
functioning. Thus a system failure, yet without any massive power component damage being inflicted.

Sooooooo, does you has either an Eee--ess--Ruh meter OR a function generator/or/audio oscillator/or pulse generator of 50 0r 100 Khz capability ? With that and your old- silly- scope, 2 toothpicks, some Scotch tape and 2 clarinet reeds, I can show you how to make a F16 screaming tree top approach, totally under the radar.

Awaiting further feedback.

73's de Edd
[email protected] ..........(Interstellar ~~~~Warp~~~Speed)
[email protected]........(Firewalled*Spam*Cookies*Crumbs)
:grin: :razz: :grin:

If you keep your feet firmly planted on the ground, you'll definitely have trouble putting on your pants.

.
The only thing I can find between the HIGH and LOW sides are two transformers as mentioned and there is ONE trace that shows schematic symbols on the silkscreen for two capacitors in series. But only one capacitor is in place jumpering the holes for the two in series. I cannot tell exactly what it's value is as it is sliver ink on a blue coating. It looks like a tantalum capacitor. Again, I cannot read what it is but the silkscreen and looks scream capacitor.

Have not traced the base legs of transistors. There do not appear to be other active components on the high side execpt the four transistors mentioned. I'm not sure of the different drive arrangements on the power transistors as I am not very familiar with fixing switchers. I've replaced fuses and power diodes in the past on computer supplies. But I have not troubleshooted a dead failure beyond that as it was way cheaper to simply salvage a computer supply out of a retired unit or buy a new supply for about $25.

There are some power diodes in the bridge/voltage doubler arrangement. I also see some small signal diodes as well but cannot recall what part of the circuit they are in and it is not in front of me right now.

This is an almost dumpster dive; and then not at all. It is a BRAND NEW refrigerator that was returned to the retail store; then the retail store donated it to the local Goodwill store. I've bought a number of "broken" things from them and fixed them. (I've bought appliances, stereos, computer scanners, toys, and so on. Normally there are some rather interesting things that get donated but then the volunteers have no idea what they are and toss them. I saw some C-Band dish stuff in the back room a while back and kept waiting for it to be put out for sale. They didn't know what it was and tossed it! I now have a deal with the manager to buy stuff like this for a reasonable donation.) This refrigerator still has the tape holding the shelves in place. They had a second one working and I can tell the two LEDs glow on the power supply board when plugged in; whether cooling or not. (I can see inside the screen/shield to where the power supply is at.) I cannot recall the date but I know it was made within the past few months; it was built in 2006.

I don't think the crash was gradual. Gradual crashes hurt as you end up switching out all sorts of caps that have dried out. I did realize I have a frequency counter that runs on 12VDC. It has an input range of 5 to 200V. I think I'll probe the base legs of the transistors with that and see if there are any oscillations. I still don't know what makes these things oscillate as it only appears to be the four transistors, the hash filter and the transformers. I could understand ONE failure of an oscillator but BOTH? I have a satellite receiver that uses a switcher. This started acting up then failed. An entire working receiver was cheaper than the power supply alone or even the parts to fix it. Thus the dead receiver is in the basement as a source of spare parts (execpt the power supply).

No function generator or pulse generator at present. I've used a computer soundcard and a program called SIG JENNY for audio testing. Used radios, cassette and CD players as well as "signal sources". (Don't knock it, they work very nicely and are "free".) Have not had the need for a true pulse generator for some time.

Hopefully I answer all the questions this time and didn't miss anything. I'm having another one of those weeks; and it is only Tuesday.
No trees were harmed in the creation of this message. But billions of electrons, photons, and electromagnetic waves were terribly inconvenienced!
dyarker
Posts: 1917
Joined: Fri Aug 22, 2003 1:01 am
Location: Izmir, Turkiye; from Rochester, NY
Contact:

Post by dyarker »

?? If the low voltage/current supply that powers the TL494CN and LM358 is out, then of course there will be no activity in the higher voltage/ high current section. The supply can't even decide if there is an under load or over load condition without logic power. I'd concentrate on low power section first.

Not as much fun, is buying a Mean-Well supply from Jameco if the objective becomes getting the fridge working.

Cheers,
Dale Y
User avatar
Bob Scott
Posts: 1192
Joined: Wed Nov 20, 2002 1:01 am
Location: Vancouver, BC
Contact:

Post by Bob Scott »

dyarker wrote: Not as much fun, is buying a Mean-Well supply from Jameco if the objective becomes getting the fridge working.
Good suggestion! I just received my first Jameco catalog in the mail last week. They seem to carry a lot of potentiometers and electromechanical stuff that Digikey does not. Products are also cheaper. Just compare the price of control knobs!

If only they had sold those Mean-Well supplies 30 years ago, I would have rolled my own high power class D amps a long time ago.

jollyrgr:

That double wound inductor on the input side is a filter for COMMON-MODE noise. It's used to block common-mode switching noise (hash as you call it) from the input. They must use something else like small shunt capacitors for reducing the differential mode noise. But you probably figured that out already.

Cheers
User avatar
jollyrgr
Posts: 1289
Joined: Thu Jan 03, 2002 1:01 am
Location: Northern Illinois
Contact:

Post by jollyrgr »

Bob Scott wrote: They must use something else like small shunt capacitors for reducing the differential mode noise. But you probably figured that out already.

Cheers
There is a square/rectangle DAIN shunt capacitor as well. Contacted HANNY and they said they'd fix it if I sent them my UPS account number and paid the shipping to them. But they would not say how much it would cost. They would not sell me the part directly, either. And this is an OEM supply apparently. I think my next step will to get a "brick" laptop supply and go there. Frig says "70 Watts" and the bigger laptops can do this easily.
No trees were harmed in the creation of this message. But billions of electrons, photons, and electromagnetic waves were terribly inconvenienced!
User avatar
jollyrgr
Posts: 1289
Joined: Thu Jan 03, 2002 1:01 am
Location: Northern Illinois
Contact:

SO CLOSE!

Post by jollyrgr »

I had time to work on this DAMN supply again. Checking parts manually I found all the power transistors, diodes, etc to be good. All symptoms pointed to a bad "startup" section. So I got a radical idea based on Edd's suggestion; a pulse generator. But NOT quite a PULSE generator. Knowing that dyarker's statement, the fact that the TL494 and LM358 didn't have power the circuit could not work, made perfect sense. I agree fully. But WHY don't these two have power? I cannot tell. Based on the FAQ for switchers http://repairfaq.ece.drexel.edu/sam/smpsfaq.htm the problem appears to be an open startup resistor.

Sooo using a combination of a pulse generator and the fact that the two chips need power, I solder a couple breakout wires from the LOW DC ground and the Vcc pin of the TL494. (I wanted easy access to the power of the PWM chip.) I plugged in the supply and of course measured zero volts on these breakout wires. Then here is where I got radical and took a chance; I connected a standard nine volt battery to the Vcc and Ground breakout wires. THE SUPPLY STARTED AND RAN! I can remove the battery and the unit stays powered on. The fans are spinning and LEDs are lit. But as soon as I unplug the power cord and plug it back in, the switcher dies. A momentary touch of the battery brings it back to life. As I type this the unit has been running for about half an hour. When fully operational the Vcc is 20 volts. As soon as the unit is unplugged and plugged back in, nothing. Just barely attach the battery and away it goes as if nothing was wrong.

So what do I look for to find a "startup" resistor? Is this a power resistor? High Ohms? Low Ohms? I've 99% beaten this dog and I cannot give up now!
No trees were harmed in the creation of this message. But billions of electrons, photons, and electromagnetic waves were terribly inconvenienced!
User avatar
Chris Smith
Posts: 4325
Joined: Tue Dec 04, 2001 1:01 am
Location: Bieber Ca.

Post by Chris Smith »

If your voltage and amperage is battery related, just use one.

Charge it like the rest of the world thru a charger.

Forget the small, nice, and neat minus all the copper, ........thats for geeks. Fix it and be done.
dyarker
Posts: 1917
Joined: Fri Aug 22, 2003 1:01 am
Location: Izmir, Turkiye; from Rochester, NY
Contact:

Post by dyarker »

A small capacitor from unregulated DC to whereever you're touching the battery to. (the smallest that starts it reliably)

When unit is plugged in, rapid rise in voltage gives current through cap to give ICs enough power to start. As a new refer, the cap may not have been put in at the factory. Among the components is there a zener to keep start up voltage to ICs from going too high? And, a reverse biased regular diode to keep voltage from reversing when unplugging?
Dale Y
Post Reply

Who is online

Users browsing this forum: Ahrefs [Bot] and 26 guests