Non inverting Schmitt Trigger question

[Dannyx]

Re: Non inverting Schmitt Trigger question

Here you go

[budm]

Re: Non inverting Schmitt Trigger question

Just curious.

[Dannyx]

Re: Non inverting Schmitt Trigger question

Any good clear hi-res straight shot of the bottom side of the board?

It shall be done. No parts there though - just traces.

[budm]

Re: Non inverting Schmitt Trigger question

Any good clear hi-res straight shot of the bottom side of the board?

[Dannyx]

Re: Non inverting Schmitt Trigger question

You do not use coil for DC current sensing, that coil at the B+ is for filtering noise from alternator just like one you have in car stereo to get rid of the whining sound from alternator.

So it seems it does not do current sensing at all, since I don't see any other large shunt element in there...

[budm]

Re: Non inverting Schmitt Trigger question

You do not use coil for DC current sensing, that coil at the B+ is for filtering noise from alternator just like one you have in car stereo to get rid of the whining sound from alternator.

[Dannyx]

Re: Non inverting Schmitt Trigger question

I found something which is interesting, but I don't think it's too much of an improvement over what I already have installed. I found this little doohickey at a flea market which claims to be a window roll-up device that supposedly does what I'm trying to build here. The price on this thing I gotta be honest was zero, because the chap gave it away as "non-functional". See, it didn't come with the wire harness (must've been pulled out of a car, who knows) and the guy said I wouldn't be able to find the exact connector to match, so he gave it away for nothing. Little did he know it's actually a standard 20p ATX connector, so I just said "sure whatever, hand it over"....he was also kind enough to include the instruction leaflet, which is mandatory for China stuff, since you'd have no way of figuring out what it is or how to use it. Anyway, enough chit-chat.

I took an ATX connector with stripped ends and wired the thing on the bench to see what it does: it behaves exactly like the one I currently have installed, except this one gives you the option of either "pulsed" operation or timed operation. If you cut the red wire, each of the relays engage for 10 seconds when it receives a lock signal and if you close that wire, each relay is pulsed very briefly, supposedly for cars which already have latching switches which only require a brief action for the window to go all the way up/down.

I was mostly interested in the control logic, I thought perhaps it has some sort of current sensing which I could hack into, but unfortunately no, it doesn't seem to do this: in "timed" mode, regardless of the load, it keeps running until the time elapses. I'm actually surprised they left the markings on the chips intact, since chinese products usually have them scratched out or use blank chips. The brains of the thing seems to be that MDT2010 chip which I couldn't find much about. The first few searches revealed a datasheet for an 8-bit microcontroller here. I thought for sure there's no way it would be the right thing, despite the model being an exact match, so I searched high and low, but all I got was this 8-bit micro....Is this thing that intelligent ? :| I mean the stuff that datasheet discusses is...intimidating to say the least :| Extremely complex stuff, so it's way over my head to try and hack this if it really IS micro-based and uses that chip indeed. The highest I'd dare to aim for at my level is Arduino There will hopefully be some engineering mavericks around here who will look at that stuff and go "oh, elementary...", so if so, please feel free to share you thoughts. Sometimes complex electronics stuff is instantly made easy when properly explained...

The other two ICs are an LM2904 op-amp and an ULN2003 relay driver, for both of which I managed to find the "correct" datasheets (I put that in quotes assuming the MDT one is indeed not right and these last ones are ). A very quick look at the circuit made me believe the op-amp is a current sense amplifier connected to that coil. I thought for sure that coil does some sort of current sensing, since it's in series with the battery input, but either I'm doing something wrong, or it's indeed not designed to do that. I also attached the instructions sheet in case you're curious or perhaps spot something that I don't. It's in "chinglish" of course, but you only really need the schematic - the other side is not that important.
What do you guys think ? Would this thing be of any use or is it going in the bin ? Cheers.

[budm]

Re: Non inverting Schmitt Trigger question

It is just Ohms law, so input the values you need to get the answer.

[Dannyx]

Re: Non inverting Schmitt Trigger question

.04 Ohms x 30A = 1.2V Vrops (more at startup). Is that a big deal?

Can I rewrite the formula as 0.04*7=0,28v ? Where 7 is an estimated current drawn after the motor has got up to speed and is moving the window freely up/down. I assume those 30a you put there were for the start inrush. There might be a small hick-up upon startup with a 1.2v loss, but should be good enough I guess...I could add a booster to allow me to use all-N channel FETs, though that would clutter up things.

[budm]

Re: Non inverting Schmitt Trigger question

.04 Ohms x 30A = 1.2V Vrops (more at startup). Is that a big deal?

[Dannyx]

Re: Non inverting Schmitt Trigger question

Back on my H-bridge project, I realised I may have some impedance issues: the P-channel FET has a 20mOhm on resistance, the N-channel has 8mOhms and the shunt resistor adds another 10, so that's almost 40mOhms of resistance which doesn't sound too good. I worry the motors might be a tad "slow"....does this usually happen with N-P channel H-bridges or is it something I don't need to worry about ? Cheers.

[budm]

Re: Non inverting Schmitt Trigger question

http://www.electronic-products-desig...neral-switches

[Dannyx]

Re: Non inverting Schmitt Trigger question

Back on this project, I wanted to go ahead and source some actual parts so we can at least prototype something, instead of just making assumptions since it could go on forever like this.

As I previously wrote, after my previous searches trying to find the "best" transistors to use (which don't actually exist, since nothing is ideal), I realized I chose too high a drain current rating for my application, but even so, I managed to get myself 2 IRF3205 N-channels...YOLO They've got a 8 mOhms D-S resistance and they probably take 100A without complaining. I only got these because they were the only ones my local electronics shop had in stock with the lowest RdsON value. Just for curiosity's sake, I also attached the list I put together before realizing I greatly overestimated the drain current, so the list is rather narrow and there might be other better choices out there, they just weren't readily available or they were expensive.

For my P-channels I found the IRF4905 in stock...the D-S resistance is a lot higher with these: 20mOhms, but again...YOLO They weren't terribly expensive and unless I REALLY screw up, I don't think I'll fry them...

[Dannyx]

Re: Non inverting Schmitt Trigger question

This might be something for you think about so hear out

How easy would it be to put a optic sensor on the window or a switch on window mechanics when the window is up and when it down

or encoder on the motor or a travel encoder on the window slide

as for as the current amp sensing to that not as easy you think and here why

is most DC amp probe that I have used you must zero it before you can use it

This could be a problem if for some reason the amp probe showing let say 5 to 10 amp or even 2 or 3 might cause you some problems

Keep mind that I am just thinking out loud

I was thinking of that solution at one point in time, however taking apart the doors, drilling holes and running extra wires is not a feasible option sadly.

[sam_sam_sam]

Re: Non inverting Schmitt Trigger question

This might be something for you think about so hear out

How easy would it be to put a optic sensor on the window or a switch on window mechanics when the window is up and when it down

or encoder on the motor or a travel encoder on the window slide

as for as the current amp sensing to that not as easy you think and here why

is most DC amp probe that I have used you must zero it before you can use it

This could be a problem if for some reason the amp probe showing let say 5 to 10 amp or even 2 or 3 might cause you some problems

Keep mind that I am just thinking out loud

[Dannyx]

Re: Non inverting Schmitt Trigger question

So leaving the relay idea behind, let's focus on the H-bridge....since there are a lot of options for N-channels and P-channels to choose from, I decided the most sensible thing would be to actually write down a list of choices which I consider suitable. However, soon after putting down around 10 options, I realised I screwed up during my searches because I went WAY overboard with the maximum drain current Id. I used the transistor tool of Mouser.com to select whatever parameters I wanted the FETs to have and then narrowed it down even further to what I'd be able to find locally, since there are obviously some very obscure, niche FETs dedicated to very specific applications which are impossible to find in my country, let alone my electronics store.

I originally selected everything 80A and up, which, I KNOW, is overkill, since even though I haven't measured the motors to know exactly their current consumption, I referred back to that article BudM provided on car accessories power figures and it estimates it to be around 30A peak. Now that I think of it, I have NO idea why I went so high initially, since I DID read that article before and even the bloody fuses are rated at 30A...lost my train of thought perhaps :| That's not to say a transistor capable of handling 80A wouldn't work, but the price is likely higher and the higher the current is, the harder the part is to source. I'd say, and please don't quote me on this, that 40A continuous Id would be enough...probably STILL too high, so feel free to correct me on this.

Also, the on resistance (RdsON) is a factor which I understand is important - the lower, the better ? The on resistance of my list of (oversized) transistors I put together before realising I chose too high a current, ranges from 4mOhms to around 15mOhms. It's also slightly higher with P-channels I reckon, but I didn't get that far yet. The shunt resistor will also add its own resistance to the mix, so I reckon I should keep RdsOn as low as possible, which is 4mOhms for an IRF1404...ok, I believe that's enough rambling. Cheers guys

[Dannyx]

Re: Non inverting Schmitt Trigger question

You do not see it?
https://www.badcaps.net/forum/attach...0&d=1499710455
Look at your diagram, if the relay contact on the right is stcuk on normally closed, and when the relay is energized, the Com left contact will make contact with the N.O. contact so you will have current flowing through the relay contacts to one stuck contact and the N.O. contact through MOSFET to GND!
Not the same as my schematic.

I believe that's exactly what I just replied, just under different words

[budm]

Re: Non inverting Schmitt Trigger question

You do not see it?
https://www.badcaps.net/forum/attach...0&d=1499710455
Look at your diagram, if the relay contact on the right is stcuk on normally closed, and when the relay is energized, the Com left contact will make contact with the N.O. contact so you will have current flowing through the relay contacts to one stuck contact and the N.O. contact through MOSFET to GND!
Not the same as my schematic, so if you hook the motor as shown, nothing will be damaged.

[Dannyx]

Re: Non inverting Schmitt Trigger question

Did you think about the failure mode of the relay contacts, I.E. one of the Contact welded closed? Nice big sparks!
Analyze, analyze......

If one of them got stuck, the motor would just have both its terminals connected to the same potential (GND or VCC), assuming the other contact DOES spring loose like it's supposed to. This holds true only for the schematic you just posted, since in my flawed application, there would indeed be a short and a blown fuse. This does not necessarily mean that an H-bridge is bullet proof and can't short out, but at least there's no mechanical failure caused by arcing.

[budm]

Re: Non inverting Schmitt Trigger question

Did you think about the failure mode of the relay contacts, I.E. one of the Contact welded closed? Nice big sparks!
Analyze, analyze......