Lc-b350atx

[Wester547]

Re: Lc-b350atx

Here are some pictures of the L&C B350 ATX completed with output capacitors replaced and sporting a much improved line filter and bridge diode. It is powering my FIC K7MNF-64 motherboard so that is where I will leave it for now. If I were to want to improve it more later, I would beef up the output diodes, especially the one on the 12 volt rail. I would also increase the lone 12 volt output capacitor from 2200 uF to 3300 uF or maybe even 4700 uF. Anyone have a recommendation for the value of the output capacitor?

You used Teapo SC to recap those failed? I hope they measured in spec.

Nevertheless, good work, as always!

[goodpsusearch]

Re: Lc-b350atx

3300uF for 12V would be enough provided it is low esr and not GP.

[everell]

Re: Lc-b350atx

Here are some pictures of the L&C B350 ATX completed with output capacitors replaced and sporting a much improved line filter and bridge diode. It is powering my FIC K7MNF-64 motherboard so that is where I will leave it for now. If I were to want to improve it more later, I would beef up the output diodes, especially the one on the 12 volt rail. I would also increase the lone 12 volt output capacitor from 2200 uF to 3300 uF or maybe even 4700 uF. Anyone have a recommendation for the value of the output capacitor?

[mikey5791]

Re: Lc-b350atx

Hi fzabkar,

Appreciate your input of maximum power rating on your LC-B250ATX circuit.
So, this tells me that the supplier label doesn't always give the actual rating.

Great sharing though.

[fzabkar]

Re: Lc-b350atx

It's basically a 200W unit. I've actually measured a power draw of 150W in the old PC that was using it.

One thing I've done with an old AT PSU was to convert it to 13.8V for powering automotive gear. I've also modified other PSUs for 6V operation. All you need to do is to change the feedback resistors in the voltage divider, and modify the OVP circuitry, if necessary.

[momaka]

Re: Lc-b350atx

^ With new caps and a few mods here and there (including a good fan set to run on 7V or so), those "crude" power supplies can provide many years of service.
Considering the fact that PCs nowadays use mostly the 12V rail to power everything, old PSUs like this would probably be only capable of running a low-end PC with a low-TDP CPU and onboard video.

[fzabkar]

Re: Lc-b350atx

It is actually quite common for Deer/L&C/Allied PSUs. Get rid of that resistor if you want your caps to live longer.

I took that supply out of service years ago. I don't expect that I'll ever use it again.

[momaka]

Re: Lc-b350atx

I am eager to find the rating on my AVF brand Model LT-35ATA PSU.

I answered your question in the thread below, so I can stay on topic here
https://www.badcaps.net/forum/showth...t=49407&page=2

The 6.8 ohm reading was a surprise.

Well, now you know. It is actually quite common for Deer/L&C/Allied PSUs. Get rid of that resistor if you want your caps to live longer.

[LDSisHere]

Re: Lc-b350atx

I am glad to see you are still resurrecting these old supplies Everell. I hope all is well with you.

[everell]

Re: Lc-b350atx

I checked my scrap pile and found a 15 ohm quarter watt resistor and a 102 1Kv disc ceramic capacitor to use in the snubber circuit. The 12 volt output diode looks burned around the leads, but measured OK. I don't have anything better to put in it at the moment, so put it back in the circuit. All parts installed. Powered it up with no mother board, and it works.

Now I will have to look in my scrap pile for some parts to make a decent line filter. The coil is wimpy and the X and Y caps are not the proper type.

[fzabkar]

Re: Lc-b350atx

Notice that the LC-B250ATX circuit has SB1040 dual diodes (10A) for the 3.3V and 5V rails. My LC-B400ATX supply has similar diodes, plus two FR302 diodes (3A) for the 12V rail.

This means that the maximum possible rating for the combined rails is ...

(3.3V x 20A) + (5V x 20A) + (12V x 6A) = 238W

The label on my supply claims that it can deliver 17A @ 12V, 28A @ 3.3V, and 40A @ 5V.

It also claims ...

+5V & +3.3V combined load 220W
+5V & +3.3V & +12V combined load 380W
Total output is 400W max.

[mikey5791]

Re: Lc-b350atx

Hi fzabkar,

Based on your earlier comment " If you calculate the actual power rating of the PSU based on the part numbers of the 12V, 5V, and 3.3V dual diodes, I expect you will find that the PSU cannot handle more than 200W." Can you explain further how to calculate the actual power rating? I am eager to find the rating on my AVF brand Model LT-35ATA PSU.

Thanks in advance.

[fzabkar]

Re: Lc-b350atx

I just measured my own LC-B400ATX:

6R8 on 3.3V
100R on 5V
220R on 12V

The 6.8 ohm reading was a surprise.

[momaka]

Re: Lc-b350atx

Is there a power supply you cannot fix?
Really nice work plugging that burned hole! The quality is up there with that repaired Delta PSU of yours.

I've seen several Deer PSUs myself with this exact problem (burning through the 12V snubber resistor after the ceramic cap shorts or becomes leaky), though none of them were this dramatic.

When you are done fixing it, make sure to re-arrange or remove the dummy load resistors as they are the primary reason those caps on the output get cooked. Once the caps get cooked, those snubber caps seems to take an extra toll with the ripple, thus burning out - or at least that's my theory, anyways.

If this is anything like the other Deer/L&C/Allied PSUs, your PSU will have some pretty low resistances on the output rails. Usually, that is a 100 Ohm resistor on the 12V rail, 47 Ohm on 5V rail, and 10 Ohms or less on the 3.3V rail. But those resistors will heat up way too much in the original configuration.

I usually do a minimum of: 270 Ohms for the 12V rail, 47 Ohms for the 5V rail, and 20 Ohms for the 3.3V rail. All resistors rated for at least 1W.

I think this PSU also has a weird loading scheme with some diodes between 5V and 3.3V rails. This can get removed without causing trouble (in fact, it will help prevent trouble and remove heat from the circuit).

Also, I suggest using polyester or metalized film caps for the snubbers. They can take a bit more abuse and are more stable with changes in temperature.

Oh, and don't forget the small 22 uF critical cap in 5VSB circuit! After all, you are the person to open our eyes behind this issue .

[everell]

Re: Lc-b350atx

A trip to the hardware store, and a little epoxy. Now the hole is plugged. Drill a couple of new holes and new parts can be installed.

[everell]

Re: Lc-b350atx

Next step was to remove the output diode, snubber resistor and snubber capacitor. The snubber resistor was burned in half. The snubber capacitor was just two wires with a little bit of burnt material on them. Obviously, the disc ceramic snubber capacitor shorted, causing all the burn damage.

Next, I scrapped all the burnt black stuff off the pc board, leaving a nice sized hole.

[Sparkey55]

Re: Lc-b350atx

She's dead- er done, Jim.

Not only is it a TO-220 device where a TO-247/262 could/should be, it's a friggin' full pack! 5A at most, for a silicon rectifier for any length of time.

"30 seconds past the 12 day warranty" doesn't count.

Snubber resistor and cap (that were?) next to it are fried too. Saw that in a clone supply, maybe Deer, where snubber cap shorted and sent ~28+VAC thru the little 10 ohm quarter watter.


"Honey, we're havin fried Deer for dinnah!"


EDIT: Nice fan control thermistor- not even in contact with heatsink... More fried Deer guts.

Most of those fan control thermistors are in the wrong place anyway. They should be in the secondary output caps area, AWAY FROM direct air flow of the cooling fans. By locating them on or near the rectifier heatsinks WITH an air stream COOLING the thermistor down will not lower the thermistors internal resistance therefore not allowing the fan control circuit to increase the fan voltage.

[kaboom]

Re: Lc-b350atx

To begin the repair process, I found and removed four bulging capacitors. Once they were out of the way, I could see a severely burned area on the pc board.

She's dead- er done, Jim.

Not only is it a TO-220 device where a TO-247/262 could/should be, it's a friggin' full pack! 5A at most, for a silicon rectifier for any length of time.

"30 seconds past the 12 day warranty" doesn't count.

Snubber resistor and cap (that were?) next to it are fried too. Saw that in a clone supply, maybe Deer, where snubber cap shorted and sent ~28+VAC thru the little 10 ohm quarter watter.


"Honey, we're havin fried Deer for dinnah!"


EDIT: Nice fan control thermistor- not even in contact with heatsink... More fried Deer guts.

[fzabkar]

Re: Lc-b350atx

If you calculate the actual power rating of the PSU based on the part numbers of the 12V, 5V, and 3.3V dual diodes, I expect you will find that the PSU cannot handle more than 200W. That's what I found for my own LC-B400ATX.

[Dan81]

Re: Lc-b350atx

To begin the repair process, I found and removed four bulging capacitors. Once they were out of the way, I could see a severely burned area on the pc board.

Yeah...that's a pretty done for 12V rectifier.

What transformer does it have? It looks almost the same as my Fortrex ST-400W (Premier LC-B400ATX) I use in my main PC. Mine has a 20A 60V 12v rectifier though. (I installed it,former was a 10A 200V rectifier )

Would be interesting to find a real ERL-35 transformer. AFAIK,some of the better L&C PSUs (like yours) used Allieds (LT35 transformer label gives it away) as their PCB,while others were the same but with usual EI-33 ASH (as opposed to the gutless EI-33 ASG ones) transformers. Mine is an example.