Видео

Inductive loads are going to respond to whatever signal you put into them. The point of the FET is to translate the low level decision making to the heavy lifting portion that makes sense for the induction (solenoids, motors, etc). If the logic level side of the FET is protected, then the load side won't notice a different. However, if levels breach or the nature of the waveform changes from what you're intending then the motor will react to the frequency content. Hope that helps!

I would just use a high side MOSFET for one application and a low side MOSFET for the other application. It's unusual to move the load's highside or lowside position dynamically in the application (there isn't a lot of reason to do so). Switching between highside power directions is the basis of H-bridges for clockwise and counterclockwise applications.

well, with the way you've communicated, it's a very generic question. If you can narrow it down for me I'd appreciate it. Usually, problems with limits tend to be about naming things. It's likely that you don't have an "OUT1" or "OUT2" node in you circuit...so MC12 would choke in that case because its trying to process something that isn't there.

If I can I will. Right now, I'm working on my house so that's been taking a lot of time.

@@pauljstar Well we certainly don't want you to have to live in your car anymore, so take your time!

Great question! I'll study and get back with you. I'm traveling now

Ok right now I see a few different things: 1. Your if statement is not enveloped by curly braces like your capacitance definition command is. I'm not certain that MC12 cares about that but it would be a good idea to make that consistent. Note: I do not know how often the conditional check occurs in processing the analysis mode you are using. Perhaps you could let me know what you're doing there. 2. The variable IndLoad is presumably an inductor part name? 3. You could implement your idea as a more real example using a idealized switch that places a capacitor onto the network for your above 300 A condition. Data about the current can be acquired by a measured change in current (sense resistor method) or use the defined symbols. If I find time I'll make a video on this, it's a good question.

@@pauljstar Thanks, but it didn't help. I'll send you an email of links to the files since RUclips won't let me post links.

Simple noise is 2 + 2 (i.e., superposition). The videos are intended for basic understanding. Your noise models can be as sophisticated as you like.

Yes, although it's called an AC analysis in the software. I was thinking of making video on that

@@pauljstar that would be great!

You're most welcome 🙏

@@pauljstar 🌺🌺🌺

Glad to help

👍

Accurate!

Glad it was helpful to you!

Happy to help!

Glad to help!

I'll try on a new video if possible. It's a bit more than what can be conveyed easily for inline on comments

I'm assuming you're asking about a direct current (DC) application and not a new concept (say alternating; AC). It would be difficult to find a single MOSFET with a safe operating range (SOA) like that. I would need to know what you're doing. That's about an 8.4 kW application. I haven't designed anything like that before.

@@pauljstar hello thanks for replying. Scratch those numbers im looking for 40 amp 90 volt dc charger with current variable.its for a e bike if you design and put it on ebay ill gladdy purchase it thanks.

You can use this to protect an inverter circuit

Yes it can! Make certain to place the zener between the correct nodes (line input and highside power supply)

100 ohms was used in this example. You are correct 100 kOhms would potentially cause the MOSFET to be slow at turn ON. 100 kOhm would form a RC network with the innate input capacitance of the gate and source.

What inverter see you using? What are you trying to do?

Typically, that's how it goes. Most applications will follow the theory-pattern you mentioned. However, current can flow either way through a MOSFET once the junction is conductive. This specialized reverse polarity protection (RPP) application uses a little trick with the MOSFET that is different than most standard applications. PRINCIPLES: A MOSFET can be understood more easily as a high impedance device when OFF and a low impedance device when ON. You'll see this tend to act like a mega-ohms resistor switching to a micro-ohms resistor. So, P-Channel and N-Channel compositions dictate the voltage arrangement by which they are switched. NEEDS: 1. The load needs a path to the supply/battery charge to operate. 2. The P-Channel MOSFET needs to activate to provide that path from the supply/battery to the load. 3. The P-Channel MOSFET needs a difference of charge concentration between the gate and source to provide the path between supply/battery and the load. PROCESS: At first, the supply/battery charge has to make it's way through the body diode to reach the source terminal of the P-Channel device. Once the charges on the gate and source terminals reach the difference threshold, the source-drain junction becomes conductive. At that point in time, the charge doesn't need to go through the body diode as much anymore. If you're curious amd want to test in the lab, you can measure OFF-state source-drain resistance directly with multimeter (regular resistance setting). You can then measure ON-state source-drain resistance indirectly with multimeter (low voltage measurement setting). Using voltage data to calculate with the known current (RDS = VGS / I). Note: The current (I) could be known by measuring it a from sense resistor, by cable clamp, or by knowing the power dissipation of a load that the MOSFET is providing access to.

I have a video on this. As in protecting the load from 50V and higher? Often you can use overvoltage protect circuits at the power supply's input. This would mean placing a pchannel in the highside or nchannel in the low side. Nchannels in the highside work also but must be enhanced by another driver circuit. This function is called "load switching" Sample the power supply input using resistor dividers (hundred kilo-ohm scale) and compare that to a voltage reference circuit (say 35 V). When the comparison proves an overvoltage event has occurred, shutdown the load switch and prevent access to the load. An analog is when someone keeps asking you for money or services...a little is ok but eventually you have to cut them off. For other kinds of overvoltage protection use a transient voltage suppressor (TVS) diode for events like electrostatic discharge (ESD). They are designed to handle the higher rate of voltage change (dv/dt) where the OVP circuit handles slower transient events.

@@pauljstar Thank you for your reply. Please can you share that circuit?

Crystal oscillator you mean right?

@@pauljstar Passive crystals piezoelectric crystal which are used in Microcontroller Thanks ❤️🙏

Anything in particular?

@@pauljstar How to probe multiple points in Schematic and view it on graph at same time

Jeez! Fused 80 A at 14 V hints at nearly a 1kW nominal application. That's why high voltage becomes really attractive to avoid waste heat. Anyway, reverse engineering is fun and also really frustrating at times! Maybe I'll do a video on it if you give me more info (that isnt NDA). The support material from Toshiba isn't really sufficient unless it's for cut and dry applications. So, doing what you're doing is in the realm of r&d experimentation (as I was). First I would investigate the feedback loops of the application. If the orginal control requires hall sensors, the TB9061AFNG cannot innately handle that (unless again, trickery). It's older brother (part number escapes me) can...just search for the hall effect sensor version if that was the case. If it is BEMF as you say, the control type is important also. It could be using field oriented control (FOC) methodology instead of squarewave... squarewave is baked into the TB9061AFNG control program. Look to see how many sense resistors are present on the board...you need at least two to use FOC methods . If there is only one sense resistor it may work. Personally, I would not use BEMF to control such a high current application. Consequences of even small control failure are blown switches (MOSFETS). You can replace them when they break but often they are applied with a massive heat sink in mind (because of high current). Typically, that's done with a lot of board real estate metal (often copper) to dissipate. So, practically you have to bake the whole board then locally desolder to reach temperatures that would yield the tab weld/adhesion of the FET package. Overall, I understand what you're doing and thinking down this track will probably get you to your reverse engineering goals. Take really good notes and collect information about every domain to learn most from each sample

I use transient most of the time because it's a good habit to look at the waveforms. Real products in the lab almost always have frequency content to observe. Simulation in purely DC framework can mislead. But otherwise yes, that function absolutely saves time that way

Really good question! For passive systems, they are nearly equivalent functionally. N-Channel applications in low side are often cheaper because manufacturers make billions of N-Channel per year (scale economy). So if you make thousands of your products saving a couple cents per device can add up. Sometimes you might not want a large volume device in the lowside/return loop area for layout reasons. That is to say, your board real estate might not resolve to something that fits the device geometry near the ground loop termination. I've done boards which it made more sense to place a reverse polarity application on the highside area compared to lowside area because it optimized the space usage better for the board. N-Channel applications require all voltages to be referenced to a separate ground node which you might not like to manage...so there's a "true ground" and "protected ground." N-Channel devices almost always have low ON-state resistances compared to P-Channel devices so N-Channels help prevent system power losses which is good. I think one disadvantage of N-Channel devices that comes to my mind would be if you have a lot of other switching circuit subsystems going on that are referenced to the protected ground. The series inductance and stray/parasitic capacitances of the N-Channel device in that ground loop position might react in a way you wouldn't expect. All the current aggregates at that N-Channel device as it exits your product. So, if the current has frequency content, it could do something weird. You'd have to look at how your switching systems interact and how their loads behave which would manifest the potential frequency content. Adding either reverse polarity device changes impedance of the lines so it's a good idea to think about how you might balance it in either application.

Perhaps I will! I like doing theoretical models first and then updating with empirical model data. The theory gives you a range of what to expect and then the lab gives you reality haha

If your simulation doesn't replicate real world conditions you are either... 1). Using a poor component model 2). Your simulation is wrong.

There is!

@@pauljstar what is where?

Thanks!

Thanks!

Great! Glad it helped!

It may be in how you're understanding electronics terminology. There isn't really ever such a thing as a driver voltage without also specifying a driver return (or often called ground). Voltage always has two points of reference and that's why it is sometimes called "potential difference." So, if the voltage between the gate and source exceeds the limit for that device, then my recommendation is to avoid encroaching or tresspassing the border. Most MOSFETs have plus/minus 20 V limit between gate and source pins. Exceeding voltage limits between two places for most devices usually ends up in some kind of dieletric breakdown effects and several orders of magnitude change in resistance (failure mode: coerced permission; such as megaohms to single digit kilo-ohms). In applications which use MOSFETs in a highside position, it's possible to have voltages of gate to the system "ground" much higher than the limit between gate and source (because the source isnt connected to ground). The MOSFET only cares about itself...so if the application space or ecosystem that it's in is friendly to the device's limits, it'll behave the way it was designed or expected to.

@@pauljstar Thanks, I do understand potential difference, however I also appreciate the additional info you took time to share. Always willing to listen and learn.

@@anondusery1271 of course, sorry if it felt weird. I never know where people are at or how they learned to speak about electronics. I taught some MSEEs while I was at ASC and you'd be surprised what crutch concepts were holding them together...so I don't like to assume anything anymore haha. Is there still something you wanted to know about gate-source limits? 🤔

You can do without a motor control predriver if that's what you're asking. However, you must choose the MOSFETs as "logic level" gated FETs. Beware of shoot through timing that would destroy MOSFET or burn motor windings.

@@pauljstar my motor 126v sir, i wat test with just 40v sir.

I'm not certain what transformer you might be talking about. It sort of depends what model you might be using. If you can send me that I can answer your question more specifically

I'll take a look and see what I can do

Thank you!

Not sure what you mean?

@@pauljstar A year ago but i guess it was the writing.

Always difficult to target exactly the concept speed because I don't have a measurable audience yet, can you suggest a topic for what you're interested in?

At the moment, I don't have a comparison video. Well the professional versions of xdxDesigner and Altium likely have a lot of preset models to choose from. However, ultimately, it comes down to one's ability to find a model from a desired manufacturer, import it, learn the parameters, tune appropriately, and use in concert with other models. Every software has idiosyncrasies with regards to that process... it just seems to be what you are willing to put up with. Free softwares almost always have more things to put up with. I like micro-cap because it's what I know well and the professional analytic features are free now.

For an igbt that might be a little different...consult the datasheet for the transistor and determine the application context a little more tightly (like what kind of control signals are actuating these devices). I don't typically use igbts but the terminal vernacular is gate, emitter, and collector... igbts are basically a mosfet controlling a bjt. They do that because mosfets are fast and bjts conduct better in higher power applications (best of both worlds kind of thing). So, anyway, try to learn more about the drive system and design from there! Good luck!

Thanks! Glad to be be a help Garry!

Great!

Yes it's called rubberbanding

It should be in the tool bar near by volatge/current/power indication toggles

@@pauljstar Ah, there it is Ctrl + Shift + R to toggle. Thank you.

Thanks I'll think I'll do that when I can