There doesn't seem to be much difference between these devices from a cursory read of the datasheets. They're all N channel MOSFETs with the same pin-outs, same max voltages, same power handling. But there must be a difference, surely. Qgs Qd Qds do vary but I'm not really sure what they related to. I'm guessing the maximum switching speed of the devices? If so, the lower the Q figures the better, right? I've got a switcher which uses an 820 which has blown, but all I've got in the spares bin are 830s and 840s, hence the question.
Yes, but that's a point in favour of substituting an 840 for an 820, all other things being equal. Just noticed I neglected to post links to the datasheets in my earlier post.
That's exactly what I was alluding to. So the higher gate capacitance will reduce the speed at which the chopper (this is a chopper FET in this SMPS) can operate, and since that speed is fixed by other circuitry, this FET will end up in some sort of permanent in-between state; never fully off nor on. Is that a fair guess?
Probably not. The bigger FET won't switch on and off as fast as the 820 so you will have more dissipation during switching, but less dissipation when it is actually on.
You could estimate both if you knew the currents and times involved, but you won't. With a dim newbie we could try to get them to think about what might be going on, but you are known to be too dim to ever get it, and far too far up yourself to ever bother.
That's mostly a a, with some help from Flyguy (who imagines his name is a reference to his claimed expertise as a glider pilot, but it seems to have more to do with his obvious pleasure in latching onto dung and reposting it here. Like your enthusiasm for Russian Today and the Daily Mail.)
So what are your estimates of the switching dissipation and the "on" dissipation of 820 and the 840 in your application?
Claiming to "get it" is what Gnatguy always does, when he clearly doesn't. Raise your credibility by giving us some numbers - "on" time versus "off" time, switching time, switching frequency. You might even be able to work out if I am right. Since you didn't give us any of this information when you posted the query, I didn't have anything to be "accurate" about - I was just giving you some conceptual tools to let you give us something we could be helpful about.
I think you got me wrong, Bill. I was genuinely appreciative of you actually providing some useful information for a change. You corrected me and gave me a very concise intuitive idea of the problem. No calcs needed. It was just the usual barbs and insults I could have done without, but hey, that's just your nature I guess and we're all well used to that on this forum. Going back to the original point, though, I was pretty much on the right lines with my supposition because it is closely related to what you said. Just a different aspect of the same issue: inability to switch fast enough and spending too much time in the ohmic region as a result (ohmic or linear region; I can never remember which is which. Anyway, the bit between on and off).
Choice of words didn't sound all that appreciative.
You hadn't spelled out your problem in enough detail for me to have posted any kind of correction.
I had to intuit what you were worried about, and spelled out a possible problem that you might have to deal with.
This isn't correction - at best potentially helpful speculation.
If you want to understand what's going on you do need to do the calculations. They don't have to be all that accurate, but you do need to do it if you don't want to blow up a lot of transistors.
More my response to your nature, which you have been telling us about here for years. A sudden burst of smarminess doesn't out years being a snarky prick.
Some of the usual offenders see a lot of it. Others don't. I play nice with people who play nice, and deal out barbs and insults to people who earn them, mainly by being rude and insulting in their responses to me.
But you couldn't articulate it clearly enough to work out what was probably going on. Finding the right point of view is frequently crucial to solving design problems, and people who can't imagine that the one they have always used isn't the right one for problem that has got them stumped are difficult to help.
result (ohmic or linear region; I can never remember which is which.
When the switch is fully on, it looks like a resistor (ohmic). When it is switching the power dissipated in the switch is the linear product of the - rapidly changing - current through the switch, and the - rapidly changing - voltage across the switch at each successive instant which you have to sum across the switching interval. That looks more like a linear amplifier, and linear audio amplifiers look like that all the time.
Integral calculus can be invoked, but numerical integration works fine, and LTSpice will do it for you if you know how to ask it to.
You need to think about what's going on in a bit more detail. If you could you wouldn't need to ask us to do it for you. Junior engineers frequently run into this problem.
There you go, you see you *can* be helpful when you choose to be. Just try to cut back on the insults and you'll be a more respected contributor here. Now, I know what you're getting at about doing the math and whatnot. THAT is an engineer's job and I'm no engineer. But I never asked for and didn't expect anyone to attempt the calcs for me and I know very well it would not be possible anyway, without data on -inter alia - the switching frequency, dv/dt, gate drive capabilities etc. The original question was only about what makes the IRF830 different from the 830 and 840. No more. It was you who inferred more into the matter than was required. Lasse provided part of the answer and you provided the other. Congratulations! Now I know you *have* to have the final word but please just try to keep it civil; thanks.
Post less fatuous right wing nonsense, and you will get a more positive reception when you do ask questions - I don't know what you think you contribute here, and your opinion of who might be a "respected contributor" isn't one that is going to carry much weight.
But even if you just want to be a hobbyist engineer, as you have claimed to be, you will get better results if work out what's going on.
the switching frequency, dv/dt, gate drive capabilities etc.
The switching frequency is something that you can see with an oscilloscope, and the switching time is visible if you hang a scope probe on the drain of the switching transistor - since you know the tie the voltage takes to swing and the voltage it swings through, that dv/dt as well,
Getting gate drive capability needs you to look at the circuit. If you plug the bits into LTSpice you can pull quite out a lot of information out of s simulation of the bit doing the switching.
Glad to be of service - even to you. Most of the questions that get asked here demand a certain amount of insight into what the poster asking the question is trying to do - dumb newbies don't know enough to tell us what they need to know. You've been posting here long enough that you should be aware of that, but you still don't know enough to ask a question that is easy to answer. The fact that Lasse didn't provide you with all the information you wanted isn't any criticism of Lasse who really does know what he is talking about, but was a trifle optimistic about your capacity to completely understand his answer (which looked pretty comprehensive to me).
It's just such a waste, because he *does* know his stuff when it comes to electronics. But we hardly ever get to see that side of him nowadays. As you may remember 25 years ago he was virtually exemplary in his contributions here. No idea why he turned into such an old sourpuss. Age I guess. I'm the same when it comes to other matters I feel strongly about. Like all the street excrement one finds in these so-called "sanctuary cities".
Well when those ocassion arise, I'm sure you'll be much more highly regarded as a result.
I've only ever claimed to be a hobbyist. I'm not sure there's any such thing as a "hobbyist engineer" as one is either an engineer and done all the exams and whatnot, or one is just a dabbler. I freely admit to being in the latter category.
Yes, criticism accepted. And Lasse is a credit to the group; always has been. Thanks again for your input.
John Larkin imagines that people who don't flatter him fulsomely are engaging in a form of ritual insult. It's as bizarre perception, but that has to be what he is complaining about.
Twenty five years ago most of the contributors were interested in electronic design - John Larkin wasn't but he did pontificate about it as if he were interested.
There are a lot more people posting here now who post the kind of right-wing nonsense that is Cursitor Doom's principal preoccupation. I slagged off people like him back then, and I'm still doing it. There are more of them now and they've put off a lot of the people who were interested in electronics
Could be - but I suspect not. The environment has changed a lot in that time.
As if Cursitor Doom and Gnatguy aren't spreading a lot of that here. Skybuck Flying is worse, but not quite so right-wing. A a is currently the chief polluter. Half the threads I see are posted by him, and hardly any of them get viewed or responded to, and those that do get responses earn them by being particularly stupid.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.