transistor LT 028G LVP 640 what is it?

[bvanevery]

I have a non-functional 200W Wagan power inverter with 3 transistors in it. On their sides they say:

LT 028G
LVP
640

I'm only able to partially decode this. I don't know what "LT" is and I've searched a lot of datasheet databases. If it's a manufacturer's code it's not a very common one. If it's an abbreviation for something I'm too much of a noob to know what it is, and it isn't in any common lists of electronics abbreviations I've found.

028 could be a part type, for instance in comparison to a MJE15028G which looks similar. Maybe the parts have TO-220-3 cases but I'm not sure. G could mean it's lead-free.

640 probably means it was manufactured in the 40th week of 2006.

I don't know what LVP is. Could be a code for where it was assembled, but if so, I don't know where I'd find such codes. Maybe one can't, without knowing who's manufacturing it, and that's the original mystery. I've read that OEM runs of components can leave you SOL as far as their naming.

As I said I'm a noob. I've used my digital multimeter's diode test function to deduce that of the 6 possible electrical flows, only 1 works, and this is true for all 3 of the transistors. If these are bipolar transistors there should be 2 flows. That would possibly indicate that the transistors are all blown in the same direction. The other possibiliy is these aren't bipolar transistors, they're something appropriate for a power inverter, and I haven't gone up the learning curve enough yet. I'll be reading how to diagnose a MOSFET shortly. It sure would help if I knew what friggin' part I'm dealing with.

If these 3 transistors are bad, then my intent is to spend a few bucks to buy some replacements, and save myself the $35 a completely new unit would cost. In which case, the new parts need to meet or exceed the specs of the original equipment. That's why I wanted to find a spec sheet for the old parts. If that's not possible, then recommendations for a "robust" replacement for a 200W power inverter would be welcome. I've seen a lot of YouTube videos of a lot of power inverter repairs, and this kind of TO-220-3 looking transistor case seems quite common in them. However I haven't seen a video or found a web/forum/blog posting specifically about diagnosing and repairing transistors in this application.

 

[davenn]

hi there
welcome to Electronics Point

LT most probably is Linear Technologies
they make many many voltage regulators etc

you didnt say how many pins the device has

may be worth contacting them with the part info and see if they can help

cheers
Dave

 

[bvanevery]

3 pins, per the possible TO-220-3 case I mentioned and the data sheet for the similar part I included. Earlier I did not find the part on Linear Techology's website, leading me to believe it's not them. Maybe it's an obsolete part they don't want to advertize, or an OEM part they did for somebody. Actually I couldn't even find something as straightforward as a transistor category on their website.

 

[(*steve*)]

As I said I'm a noob. I've used my digital multimeter's diode test function to deduce that of the 6 possible electrical flows, only 1 works, and this is true for all 3 of the transistors.

That is somewhat indicative of a power mosfet. Beware that these have a very sensitive gate which can be destroyed by an unnoticed tiny static charge when you pick it up.

If it is a mosfet, the gate is the terminal that is not involved in that diode action you see. Get a 9V battery, with a 1k resistor in series with it, and connect it between one of the two other terminals and the gate. Some combination of polarity and pin will cause the terminals you identified earlier to conduct in both directions.

Once you've done that, the source terminal is the one the battery is connected to (beside the gate) and the drain is the one that is not connected. The tab of the TO-220 is normally connected to the drain if it is connected to anything. (also the tab is typically connected to the middle pin, so that is also often the drain).

My guess is then that you see conduction when the multimeter is across the middle and right leads. try connecting the battery first between the outermost legs.

If the device turns on with the +ve lead connected to the gate, the device is an N channel mosfet. If the -ve lead is connected to the gate it's a P Channel mosfet.

Note that the device will stay turned on (or off) even after you remove the battery. It's a good idea to keep all three leads shorted together when the device is out of the circuit so your handling doesn't damage it. Wrapping it up in a small piece of alfoil is a way that uses easily available household stuff

If these 3 transistors are bad, then my intent is to spend a few bucks to buy some replacements, and save myself the $35 a completely new unit would cost. In which case, the new parts need to meet or exceed the specs of the original equipment. That's why I wanted to find a spec sheet for the old parts. If that's not possible, then recommendations for a "robust" replacement for a 200W power inverter would be welcome. I've seen a lot of YouTube videos of a lot of power inverter repairs, and this kind of TO-220-3 looking transistor case seems quite common in them. However I haven't seen a video or found a web/forum/blog posting specifically about diagnosing and repairing transistors in this application.

It really helps to know what voltages and currents are involved, and also the driving capability of the circuit.

Just to make things more complex, there are three terminal devices which look and test like mosfets, but which are significantly smarter, although I wouldn't suspect you have one of these.

Can you post a photo of the circuit board so we can take a look?

Please tell us what voltage this operates from and the current it draws (or the fuse rating if you're unsure of the former)

 

[bvanevery]

I'll do the testing tomorrow when I'm more awake. Meanwhile I'm in a "use it or lose it" situation on a BestBuy gift card, so I think I'll be ordering a replacement power inverter. I wouldn't mind continuing to bang on this one though, as it would be good to have a spare if I can fix it, and it's educational if I can't. Meanwhile...

Note that the device will stay turned on (or off) even after you remove the battery. It's a good idea to keep all three leads shorted together when the device is out of the circuit so your handling doesn't damage it. Wrapping it up in a small piece of alfoil is a way that uses easily available household stuff

...nothing's leaving the circuit board unless it's determined to be dead. In which case the point of saving it is moot. It is entirely possible that these presumed MOSFETs are just fine.

Can you post a photo of the circuit board so we can take a look?

When I'm wakeful. Meanwhile,

Please tell us what voltage this operates from and the current it draws (or the fuse rating if you're unsure of the former)

12V DC in, 20A fuse, 200W AC out at standard US frequency.

 

[bvanevery]

Get a 9V battery, with a 1k resistor in series with it, and connect it between one of the two other terminals and the gate.

This happens to be a futzy step for me, as I don't have 1k resistors lying around. What happens if I just plug the inverter into my 12V car battery like it's designed to be used? Are these presumed mosfets likely to change state then? Will using my multimeter on the circuit while it's plugged in, but there's nothing loading it on the AC side, be bad in any way?

 

[(*steve*)]

...nothing's leaving the circuit board unless it's determined to be dead. In which case the point of saving it is moot. It is entirely possible that these presumed MOSFETs are just fine.

In that case you may not be able to identify the part, nor determine if it is faulty.

But that's your choice.

I'm just indicating some of the precautions you might need to take if you remove it, and how to test it once (and if) you do.

When I'm wakeful.

Great, we probably shouldn't go any further until you can post them.

12V DC in, 20A fuse, 200W AC out at standard US frequency.

OK, so 110V RMS and I bet it's not designed for 200W continuous... It would be getting very close to the rated current for that fuse assuming a normal efficiency.

Do the specs say infighting like "square wave", "modified square wave", or "sine wave" on them? Or possibly a link to information about the inverter on the net?

I'd be almost 100% sure it will be a fairly simple design, but it may be worth confirming that (the switching speed of the transistors will need to be much higher for a device that creates a sine wave output as it will be doing PWM. Square wave output just needs to switch on and off at the line frequency.

This happens to be a futzy step for me, as I don't have 1k resistors lying around. What happens if I just plug the inverter into my 12V car battery like it's designed to be used? Are these presumed mosfets likely to change state then? Will using my multimeter on the circuit while it's plugged in, but there's nothing loading it on the AC side, be bad in any way?

The inverter is dangerous and at present I would not suggest you apply power with the case open.

The test with the battery and the resistor only apply once you've removed the transistor anyway.

 

[bvanevery]

I'm slow about the photo because I've never quite squared that on Linux, and also the point is starting to seem moot.

OK, so 110V RMS and I bet it's not designed for 200W continuous...

It wasn't a casual purchase 4 years ago, it was meant to do a serious job in the woods, so yes it is continuous.

It would be getting very close to the rated current for that fuse assuming a normal efficiency.

200W / 12V = 16.6A. I wouldn't consider a 20A fuse to be "very" close. I haven't put equipment on it that would load it that high anyways. A 90W laptop power brick and a 50W external battery charger. I suspect death by other means. I did test the fuse in the power inverter, it's fine.

Do the specs say infighting like "square wave", "modified square wave", or "sine wave" on them?

At this build and price point it is of course a modified sine wave inverter. Mine is an older rev of the Wagan 2402-5. I just ordered a new 2402-5. After reading lots of power inverter reviews, I couldn't find anything better at BestBuy online, and I'm not sure there's anything better at that price point anyways. Probably they're all between somewhat ok and complete junk, and at least this one's quiet. No fan.

The inverter is dangerous and at present I would not suggest you apply power with the case open.

Good point. Funny how I miss the simple things.

The test with the battery and the resistor only apply once you've removed the transistor anyway.

That's an impasse then. There's not much point in pulling the presumed mosfets off when I don't know that they're the problem and I've ordered a replacement unit anyways. I am somewhat curious to know if the new unit has exactly the same components in it, or if they've been revved to something else.

My inability to identify those codes irks me.

 

[(*steve*)]

200W / 12V = 16.6A. I wouldn't consider a 20A fuse to be "very" close.

If you assume the device to be 90% efficient (and that's pretty high) then you get 18.5A. If you allow for the batter voltage falling to 11.5V then you get to 19.32A

Yep, I'd call that very close.

Of course, if your engine is running and the system voltage is closer to 13.8V then the current will only be 16A.

 

[bvanevery]

Hm, well, real conditions I've measured could have been 11V on the deep cycle battery at the absolute lowest, and probably more like 11.8V as 11V was on the pretty much dead side of things. Real load should never have been higher than 140W as those were the devices plugged in. Unless both of those devices can spike, which I very much doubt, since one is only a laptop and the other is only an external laptop battery charger. The laptop usually claimed to be consuming 25W, not 90W, although occasional heavy 3d use could gave gotten it to 90W. Not my typical usage pattern though as I preferred battery life for the most part. Also if I was doing heavy 3D I wouldn't have been charging the external battery at the same time, so it never would have been at 140W. Even if I had, 140W / 11.8V = 11.86A, nowhere near the 20A fuse capacity. The fuse never blew, I tested it, and I even swapped "known good" fuses on someone's theory that a fuse can look right but not be right.

Death by other means. I wonder if using a bad AC power brick did some kind of backfeed to kill it. I did have some crackling going on in one of my AC bricks before it died.

Maybe it's time to start posting in a section about diagnosing power inverters, since I totally struck out as far as identifying the part codes.

 

[bvanevery]

what killed my 200W power inverter?

I have a dead Wagan 2042 200W power inverter. I bought it in 2010. This is an older version of the current Wagan 2042-5 product. The inverter was used with an Odyssey deep cycle battery in my car, a drop-in replacement for a typical SLI battery.

It was used to run a laptop with a 90W power brick. The laptop rarely drew that, as I had it on low power settings to preserve battery life in the woods. The laptop typically claimed to be using 25W and only heavy 3D use would get it to 90W, which I never really did. Fudging to double, let's say the laptop drew 50W in the real world. I also at times plugged in an external laptop battery charger, a 50W rated device. Total draw 50W..100W, well below the inverter's capability. The inverter has a 20A fuse which never blew. I checked the fuse with my multimeter, it's fine, and I also tried known good fuses just to be sure.

The inverter had a previous death for about a year. Eventually I got bold and opened up the DC power plug. The solder on the wiring to the negative terminal had broken. I wrapped the bare wire around it with electrical tape and everything worked fine for several years after that. Recent continuity tests showed nothing wrong with the wiring, although I could resolder it for sake of form.

Near the time the inverter died, I was using some bad AC bricks. One of 'em would crackle. I thought perhaps that was from some bare wiring where it went into the AC brick. I'd try to tape it but it would always come off. Eventually the AC brick died. I also distinctly remember a low volume but high pitched electrical whining sound coming from something, whether from the inverter or the dying AC brick, I'm not sure.

Opening up the inverter, I see no burn marks, no blown capacitors, no obvious signs of failure. I've watched a number of power inverter repair videos and have done tests on the AC plugs. I don't find anything shorted. Power switch is fine. I don't see any melted or bad solderiing jobs on the back of the board.

The inverter has 3 trasistor-looking devices in it. A diode test shows that of the 6 possible electrical flows, only 1 has a voltage. If they are bipolar transistors, that would mean they're all similarly blown. But if they're mosfets, they could be working as designed and not the problem at all. Outside chance they could be ASICs in that kind of packaging. The code on the side of those components is inscrutable:

LT 028G
LVP
640

I brought that up in the identification forum and nobody could identify it. LT isn't any obvious manufacturer code. In particular it's not Linear Technologies, or at least, they have no such part on their public website. Maybe they OEMed it and I wouldn't be able to know.

The "presumed mosfets" seem to handle the AC out functions. One has the blue positive wires soldered to it, going to the positive part of the AC plug. One has the brown negative wires soldered to it. One just has a resistor, and there are no grounding wires on this device. The ground plugs are just dead holes with nothing in them. The ground plug mosfet shows asymmetrical resistance compared to the positive and negative mosfets, but that could just be its design. The positive and negative ones are attached to wires, the ground one is attached to a resistor.

So, what killed old Bessy? Just one of those tiny surface mounted parts gave up the ghost? Planned obsolescence? Feedback from the dying AC bricks? What kind of damage was likely done?

Is it worth bothering with replacing those mosfets to try to repair this thing? I already ordered a replacement inverter because I had an expiring gift card I had to use up, so this would be about having a spare.

As far as educational value, I've gotten a good amount already. I've learned that identifying a weird code is a PITA, and that the differences between transistor-like devices are a PITA. I could read up more about diagnosing mosfets on a circuit board. Another fellow suggested some off-board tests for their health, but I'm not pulling them unless I know they're bad.

 

[bvanevery]

What if "LVP" stands for Low Voltage Programming and these are some kind of Programmable Integrated Circuit (PICs) ? Does that change one's strategy for identifying the friggin' part? If one would still look in all the usual databases of the usual datasheets, the part still doesn't come up. Linear Technologies still doesn't promote any knowledge of the device, does not identify a "028" or "028G" on their public website.

 

[Jagtech]

Sorry to hear that your "old faithful" inverter has finally died. Those inverters are quite conservatively rated, with a peak power rating nearing 500 watts.
That said, it is a MSW modified sine wave inverter, as opposed to a pure-sine inverter.
That may explain why your AC brick complained when plugged into it, as SOME ac bricks don't like modified sine-wave input voltages. The brick likely failed due to this, and overloaded the MSW inverter. Keep this in mind if looking to replace Bessy.
In the meantime, most times (not always, tho) mosfets that fail in an inverter will measure as shorted, or very low resistance. You likely have several blown mosfets, which if replaced, may rectify the situation.
The LVP640 is a pretty standard 200w N-channel mosfet, quite cheap. I would look at those first.
Good luck!

 

[(*steve*)]

The power factor of the power supply you were trying to drive was such that it would draw high peak currents at or near the max voltage.

This would have (most likely) exceeded the current capability of the transistors and killed the unit. (and in your case it took aout a few of the power supplies before one of them finally succeeded in doing what the others had failed to do).

I would recommend you get something with a higher capacity (500W?) to operate that SMPS. Even better, get a power supply for your laptop that runs on 12V.

 

[bvanevery]

Jagtech, how is it that you were able to identify those codes in 5 seconds, and nobody in the parts identification forum was able to? Just the luck of who's interested in answering or bored with a post? Based on an inscription style for a different part, I thought maybe LVP was some kind of manufacturer code, and 640 was a date code. So if LVP640 is what the gizmo actually is, what the heck is LT 028G?

I'm not understanding "power factor". It's a 90W Dell AC power brick. Are the specs on the brick pretty much lying, how far of a draw over 90W is it going to go? The wiring from the cigarette lighter to the battery can only handle 20A anyways and that's what it's fused for.

I do have a DC power brick for that Dell. The problem is, it wasn't cheap, and Dell has this 3rd wire to narc on whether you're using a Dell charger or not. If you're not using a Dell charger, it will throttle your power and refuse to charge your battery. There is no BIOS workaround because it's a straight embedded circuit determination. Eventually that 3rd wire fries somehow. Either the wire itself gets bent, or the component providing the embedded device code goes bad. So I have a DC power brick that will provide power to my laptop but won't charge it. Its cord is also rather short and DC extension cords had too much loss to be practical in the car.

Hopefully we will soon have 100W USB power as standard and the proprietary charger wiring issue goes away.

My experience is that recent Chinese knockoff AC power bricks cost $8 and last 1 year. That's pretty good considering Dell used to charge $60 for such things and you'd only get 2..3 years out of them.

 

[Jagtech]

Jagtech, how is it that you were able to identify those codes in 5 seconds, and nobody in the parts identification forum was able to?

40+ years in the electronic repair business....

 

[davenn]

bvanevery

I have merged your 2 threads as they are on the same topic

dont start new threads for the same subject

Dave

 

[bvanevery]

I beg your pardon, my 1st thread started out with the narrow goal of identifying the part number, which is why I posted in the DSMP subforum. Nobody was able to answer my question. The thread started to drift towards underlying causes of failure, which isn't topical to DSMP.

So I posted a new thread in Electronics Repair subforum. I gave the complete history of the device, which i didn't before. Surprisingly I got the answer about part identification in *that* forum. Crucial information I otherwise would not have obtained around here. And I got it by following a topicality imperative.

I will do it again when confronted with the same evolution of subject matter. If you need to ban me at that time, do so. I'll find somewhere that people concentrate on the intentions and the results.