Keyboard Shortcuts
Likes
- LTspice
- Messages
Search
|
Re: plotting the difference of voltages
John Woodgate
In message <CAK-wn_46VVk4kRHPLiArsHmsDFgdzV=ec+1xnWs1oo86_Q29oA@...>, dated Sun, 25 May 2014, "Sam Jesse revrvr@... [LTspice]" <LTspice@...> writes:
In Files/Temp/ transformer.ascYou can easily see that V1 is sometimes a positive voltage and at other times a negative voltage, while V2 is opposite. So (V1-V2) is sometimes a positive number and sometimes a negative number. Think about the 'number line', which has zero in the middle, positive numbers going off to the right and negative numbers going off to the left. Then consider adding +3 to -2 and subtracting, first +4 from +2 and then +2 from -3. -- OOO - Own Opinions Only. With best wishes. See www.jmwa.demon.co.uk Nondum ex silvis sumus John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK |
|
Re: SMPS switching losses in LTspice?
The lack of such information suggests that the transistor might not be a good choice for this use.
toggle quoted message
Show quoted text
Jim Wagner Oregon Research Electronics On May 24, 2014, at 4:51 PM, sawreyrw@... [LTspice] wrote:
|
|
Re: SMPS switching losses in LTspice?
potrstuvich,
The data sheet specifies the gate drive voltage, Rg, Id and Vds.? A MOSFET is almost always specified with an inductive load which means the load current is essentially constant during switching.? Therefore, your test circuit should use Rg and Vg in the gate and a current source connected to the drain.? The drain voltage should be clamped to a maximum voltage of the Vds spec.? Regarding Cgd in the model, you should be able to get a sense of whether the model is reasonable given the data sheet value.? Someone else suggest you should also look at the Qgs spec, and I agree. The net net is that you can build a test circuit to get a reasonable engineering estimate of the "goodness" of the model. Rick |
|
Re: SMPS switching losses in LTspice?
BSC159N10. its datasheet gives no switching tests with which I can simulate and compare. So there is no way of me knowing whether or not the 1.07W of switching loss I get for this model is accurate. This fet is the one provided by the ltspice program. If you click it, the model just says cgd(min) = 15p and c(gd)max = 450p. ? I am going to have to conclude that I don't know whether or not the switching losses in ltspice for this particular model are accurate enough. I understand? that ltspice is capable of doing this, but don't know how to write the model statement for that BSC fet, and dont know how to analyse the model statement for the fet in the ltspice program. |
|
Re: Analog multiplier - wrong simulation or flaw in the circuit?
Hello Augustinto,
There are two problems in your circuit. 1. The offset adjust circuit only works with a very few hundred milli Volts. See my test circuit. 2. Your Opamps require a feedback capacitor of a few ten pico Farads. The folder with the circuits: ? It's also very important to use matched transistors.1. Use matched quad transistors! MAT04, MAT14, THAT300, 2nd choice CA3046, CA3086, CA3127 2. Don't forget decoupling capacitors of > 0.1uF from +V to GND and -V to GND and a real circuit! Best regards, Helmut |
|
Analog multiplier - wrong simulation or flaw in the circuit?
I run across the schematic shown in page 13 of the RC4136's datasheet (uploaded). After 3 days trying to have it "working" in LTSpice (uploaded) I realized I need help.? Sorry but I am not sure if I uploaded both files (RC4136 datasheet and Analog mutiplier 03.asc in the right place) |
|
LM2917 F/V Tachometer simulation problem
A few days ago, 'hitec92407' uploaded the file "LM2917Test.zip" to the group's Temp area. ?Through my mistake, the discussion ended up on the [LTspiceFiles] group rather than this one.
hitec92407's questions included these: He created a dual-collector transistor model. ?He was unsure about it. My response: It looks good to me. ?I think you did it right and I don't think it is why the simulation doesn't work. He noticed the output of the tach circuit always stays the same and doesn't do what it should. My responses follow.
The circuit, taken right out of Figure 14 in National Semi's Applications Note AN-162, is peculiar because the on-chip "ground" (VSS?) pin is not grounded. ?It is connected to a point that is nominally half-way between power and actual ground. ?That means some of the signal pins are well below on-chip VSS. ?Usually the on-chip VSS is the substrate, and nothing should ever be more negative than that. ?It just doesn't seem right.
I have a strong suspicion that Figure 14 might not actually work.
In addition, there might be problems with the IC model too, in spite of the fact that it looks like a faithful replica of their Figure 2. ?The on-chip regulator is a few volts off. ?The input pin has a 10K series resistor, which seems to prevent the input from noticing zero-crossings (i.e., it's impossible to bring pin 1 low enough to ever toggle the Schmitt trigger ... so it never sees any AC input!). ?There may be issues with mismatch between diodes and transistors, which prevents some transistors from ever turning on. ?Those are the problems I saw so far.
I can see nothing happening in the Charge Pump. ?Its input seems to be forever stuck high. ?The problem starts in the input hysteresis amplifier.
It might be possible to "patch up" your LM2917 schematic and make it work, but there may be a better approach. ?There already is an LM2917 LTspice model in this group's Files section. ?It is not in schematic form, and I can give no guarantee about how accurate it is, but apparently someone found it useful enough to upload it here. ?Try it out.
Regards,
Andy |
|
Re: SMPS switching losses in LTspice?
potstuvich,
I don't think it's a question of how accurate is LTspice, but instead, how good is the transistor model?? If I really wanted to know the answer to this, I would build a switching time test circuit that matches that used in the data sheet and see how the results compare to the data sheet.? I realize there is not much information on the data sheet, but it's better than nothing.? I would also compare the data sheet capacitancs to the model. Rick |
|
Re: SMPS switching losses in LTspice?
I would expect it to be modestly accurate, then.
toggle quoted message
Show quoted text
Jim Wagner On May 23, 2014, at 1:19 PM, potstuvich@... [LTspice] wrote:
|
|
Re: SMPS switching losses in LTspice?
Should have written, instead of "losses will be mostly resistive" it ought to be "inductor losses will be mostly resistive" since gate charge losses CAN be significant.
toggle quoted message
Show quoted text
Jim Wagner Oregon Research Electronics On May 23, 2014, at 12:54 PM, Jim Wagner wagnejam99@... [LTspice] wrote:
|
|
Re: SMPS switching losses in LTspice?
Those voltage dependent FET capacitances are modeled well, if a good model is supplied, For an external FET switcher, that is up to you.
toggle quoted message
Show quoted text
The switching loss estimate is as good as your model. For example, if the inductor stays well away from saturation, then the losses there will be mostly resistive. If you have a reasonable series resistance in that model, then its good. But, if it is driven into saturation and you do not have a model that includes saturation, then the loss number will be overly optimistic, maybe by a lot. Jim Wagner Oregon Research Electronics On May 23, 2014, at 12:43 PM, potstuvich@... [LTspice] wrote:
|
|
SMPS switching losses in LTspice?
Hello, LTspice calculates a switching loss of 1.07W in the FET of the buckboost smps that I have uploaded to the Temp Files area. "Buckboost SMPS _switching loss.asc". How accurate is this to the real circuit? I mean, the switching Mosfet capacitances, especially Cgd, are voltage dependent, and this would need modelling to make the switching losses accurate. How accurate is it in LTspice? |
|
Re: Pull-Push transformer for valve amp in LTspice
¿ªÔÆÌåÓýTip:Ferrite cores are identified by an Al number, which allows quick determination of the number of turns, using the formula L=Al.N? Le 23/05/2014 15:15,
pha0001@... [LTspice] a ¨¦crit?:
? |