I love LTSpice. I love the way it works flawlessly under wine. After all these years they have not used any Windows features that would cause wine problems. Sometimes I suspect they are actually being careful not to do that.
I always though LTSpice was just a GUI front end bolted onto some existing Spice engine. I recently found out it's actually a highly optimized Spice engine written by Mike Engelhardt who seems to have spent his whole career so far building it! If you're interested you can hear the story in this interview http://www.theamphour.com/196-an-interview-with-mike-englehardt-spice-simulator-synteresis/
That Pulsonix may be a wonderful simulator but wow, it's closed source, it's for legacy operating systems only, they don't even tell you what it will cost so guess it's far to much to think about. They use "dongles" for license enforcement (Is this the 1980's?).
Sadly unusable.
Edit: On the other hand....It seems the SIMetrix SPICE free demo runs fine under wine on Debian. Looks a bit clunky compared to LTSpice and it limited to 140 nodes. I'll give it a little check out.
I am just curious about the databases. It seems to me that any circuit simulator is only as good as the database for all the components you want to include.
I did notice that several free alternatives in Linux were very short of data.
LTSpice at least has a ton of component models. Perhaps not exactly the transistor or opamp you want but usually you can find something close enough to be able to explore your circuit ideas.
Then of course you can always scour around the net for models (No not that kind of model...). I'm sure component vendors have models of their devices for potential customers to use.
I forget the details but I believe there are couple of versions of the Spice model syntax and you need to get the right one or know how to adapt them.
A few years back I had models for vacuum tubes in LTSpice working...
I tried the SIMetrix free demo once, and didn't like the schematic editor much. I was used to the Pulsonix editor, of course, which is used for their version, and is very slick.
Does anyone know of any other good learning materials for those who are just getting into circuit design? I am still a newbie when it comes to circuit design, but I would like to get more experience and learn how everything works together to form complicated circuits.
Does anyone know of any other good learning materials for those who are just getting into circuit design? I am still a newbie when it comes to circuit design, but I would like to get more experience and learn how everything works together to form complicated circuits.
A circuit simulator does not help you learn anything about circuit design.
Yes you could probably figure out which symbols in the simulator correspond to what components in the real world. You could probably figure out how to wire them up and run simulations. Then what? Are you going to spend the rest of your life trying all kinds of circuits and component values until you find something that does what you want?
No, what you need is a good book or tutorial. The all time classic text is "The Art of Electronics" by Horowitz and Hill. http://www.amazon.com/exec/obidos/ASIN/0521370957/ It's a tad expensive and I'm waiting for the new edition which was slated to be out about now...
its over 1000 pages and earlier I was looking up transistors ,a good 30 pages or more. I just want to switch something using a breadboard. no reson except education. is there an easy way looking which pins are collector/emitter ?
When studying a circuit in a book or elsewhere it's great to be able to build it and play with it a measure things. Then the ideas are more likely to stick in your brain.
Of course that can be a lot of work and expense so messing with the circuit on a simulator is the next best thing.
You can teak around with component values and different configurations, "measure" what happens and convince yourself that it really does behave the way the theory said it might.
Be careful. I don't know for sure about the 3rd edition but the second edition has an "errata" available that's over 100 pages. The book is so full of mistakes that it should have been recalled.
When studying a circuit in a book or elsewhere it's great to be able to build it and play with it a measure things. Then the ideas are more likely to stick in your brain.
Thx, I'll be making some very basic circuits to start with !
Be careful. I don't know for sure about the 3rd edition but the second edition has an "errata" available that's over 100 pages. The book is so full of mistakes that it should have been recalled.
Well i'll hope they got it right in the third edition. the part I read about transistors was well explained, putting some of those diagrams into a simulator would help and make it stick !
LTspice is definetly a good software in terms of ease of use and amount of effort linear puts into providing support for it, but it has two main disadvantages:
1. Damn slow, even on modern octa-core CPU's
2. Software easily passes the impossible parameters. For example, it is possible to deliver absurd current and voltages using standard components, which won't work in real life (say using 1 meg and 0.01 ohm resistors in feedback divider).
Software easily passes the impossible parameters. For example, it is possible to deliver absurd current and voltages using standard components, which won't work in real life (say using 1 meg and 0.01 ohm resistors in feedback divider).
That is not a a deficiency in LTspice. Any professional sim will provide the results you ask for without making a judgement call. The user is required to have a brain and some level of intuition and common sense.
Damn slow, even on modern octa-core CPU's
Post that on the LTspice forum with an appropriate example circuit and wait for the response
For example if you draw a typical bistable (google it) with two transistors and four resistors you may well find it does not flip into one state or flop into another but rather it will get stuck in the middle. I real circuit would never do this. You can make one collector resistor a tinty bit different from the other and then it will flip or flop correctly.
Thing is with all the component models being perfectly similar on both sides of the circuit it cannot decide which way to flop. It's a bit like have a physics simulation of a perfect cone standing on perfectly upright on its point. The simulation cannot decide which way it should fall. So it balances there forever. In reality a cone would always fall down.
Then you can have fun with circuits that generate a billion giga volts or consume a billion giga amps. And nothing explodes!
Having a brain may or may not help but it would be a lot more fun if in simulations electrolytics would explode when wired backwards or resistors vaporize when asked dissipate a megawatt and so on. It's time we had a Spice simulator integrated with something like an Unreal game engine. It's a shame someone like Michael Carmack isn't into electronics:)
I have just tried Yenka for circuit simulation. the best part is the many tutorials and examples. just a quick look have to go to get my jaw fixed at the dentist in half an hour. oh boy I hope I feel better when its over !
Having a brain may or may not help but it would be a lot more fun if in simulations electrolytics would explode when wired backwards or resistors vaporize when asked dissipate a megawatt and so on. It's time we had a Spice simulator integrated with something like an Unreal game engine. It's a shame someone like Michael Carmack isn't into electronics:)
Heater! I'm surprised, you've actually come up with a good idea!
I'm now visualizing something like Xbox Kinect (or whatever) where we can grab and plug in virtual components, whoops, wrong way round. Oh no, there's A.I. to finish me off and I can smell smoke, it's the PC, quick get the fire extinguisher! This VR is Very Real indeed!
Heater! I'm surprised, you've actually come up with a good idea!
I'm now visualizing something like Xbox Kinect (or whatever) where we can grab and plug in virtual components, whoops, wrong way round. Oh no, there's A.I. to finish me off and I can smell smoke, it's the PC, quick get the fire extinguisher! This VR is Very Real indeed!
One of the best books I found quite by accident at a Goodwill Store: "Introductory Electronic Devices and Circuits" by Robert T. Paynter c1991. Each chapter ends with a section on troubleshooting devices and circuits just covered which made it a rather unique find. If you are a real glutton for punishment, you can find a Millman and Halkias edition archived somewhere or a Tektronix Series on Semicoductor electronics in bite size training mode from 1960's I think. Oh, you may be able to find the Navy's Basic electronics series online as well. Have not looked at it in quite a while. First time through was in 1977. Before electrician "A" school.
The sims are an okay way to sanity test, but without the underlying knowledge, you will not have an understanding of what you are doing, to say nothing of what is really possible. You will be shackled to other peoples designs and will likely be unable to proceed any further when the working sim fails in reality. Frustration at best. Play sure, but build the knowledge.
Comments
That voltage boost circuit I sent you should look like this once you run it and add the probes (you can figure that part out).
I always though LTSpice was just a GUI front end bolted onto some existing Spice engine. I recently found out it's actually a highly optimized Spice engine written by Mike Engelhardt who seems to have spent his whole career so far building it! If you're interested you can hear the story in this interview http://www.theamphour.com/196-an-interview-with-mike-englehardt-spice-simulator-synteresis/
http://www.simetrix.co.uk/site/index.html
It is supplied with lots of examples.
That Pulsonix may be a wonderful simulator but wow, it's closed source, it's for legacy operating systems only, they don't even tell you what it will cost so guess it's far to much to think about. They use "dongles" for license enforcement (Is this the 1980's?).
Sadly unusable.
Edit: On the other hand....It seems the SIMetrix SPICE free demo runs fine under wine on Debian. Looks a bit clunky compared to LTSpice and it limited to 140 nodes. I'll give it a little check out.
I did notice that several free alternatives in Linux were very short of data.
LTSpice at least has a ton of component models. Perhaps not exactly the transistor or opamp you want but usually you can find something close enough to be able to explore your circuit ideas.
Then of course you can always scour around the net for models (No not that kind of model...). I'm sure component vendors have models of their devices for potential customers to use.
I forget the details but I believe there are couple of versions of the Spice model syntax and you need to get the right one or know how to adapt them.
A few years back I had models for vacuum tubes in LTSpice working...
I tried the SIMetrix free demo once, and didn't like the schematic editor much. I was used to the Pulsonix editor, of course, which is used for their version, and is very slick.
Thanks for the help.
A circuit simulator does not help you learn anything about circuit design.
Yes you could probably figure out which symbols in the simulator correspond to what components in the real world. You could probably figure out how to wire them up and run simulations. Then what? Are you going to spend the rest of your life trying all kinds of circuits and component values until you find something that does what you want?
No, what you need is a good book or tutorial. The all time classic text is "The Art of Electronics" by Horowitz and Hill. http://www.amazon.com/exec/obidos/ASIN/0521370957/ It's a tad expensive and I'm waiting for the new edition which was slated to be out about now...
Other than that a google for "electronics tutorial" will get you a ton of good stuff. Like http://www.electronics-tutorials.ws/
its over 1000 pages and earlier I was looking up transistors ,a good 30 pages or more. I just want to switch something using a breadboard. no reson except education. is there an easy way looking which pins are collector/emitter ?
I gues I didn't understand that posters question
nice link http://www.electronics-tutorials.ws/
and ill probably use ltspice to make circuits I see in books and the web.
When studying a circuit in a book or elsewhere it's great to be able to build it and play with it a measure things. Then the ideas are more likely to stick in your brain.
Of course that can be a lot of work and expense so messing with the circuit on a simulator is the next best thing.
You can teak around with component values and different configurations, "measure" what happens and convince yourself that it really does behave the way the theory said it might.
Here's an example: http://www.partsim.com/simulator/#19974
Be careful. I don't know for sure about the 3rd edition but the second edition has an "errata" available that's over 100 pages. The book is so full of mistakes that it should have been recalled.
Thx, I'll be making some very basic circuits to start with !
Ive just been having a quick go at, nice !!
Well i'll hope they got it right in the third edition. the part I read about transistors was well explained, putting some of those diagrams into a simulator would help and make it stick !
:coffee:
1. Damn slow, even on modern octa-core CPU's
2. Software easily passes the impossible parameters. For example, it is possible to deliver absurd current and voltages using standard components, which won't work in real life (say using 1 meg and 0.01 ohm resistors in feedback divider).
That is not a a deficiency in LTspice. Any professional sim will provide the results you ask for without making a judgement call. The user is required to have a brain and some level of intuition and common sense.
Post that on the LTspice forum with an appropriate example circuit and wait for the response
For example if you draw a typical bistable (google it) with two transistors and four resistors you may well find it does not flip into one state or flop into another but rather it will get stuck in the middle. I real circuit would never do this. You can make one collector resistor a tinty bit different from the other and then it will flip or flop correctly.
Thing is with all the component models being perfectly similar on both sides of the circuit it cannot decide which way to flop. It's a bit like have a physics simulation of a perfect cone standing on perfectly upright on its point. The simulation cannot decide which way it should fall. So it balances there forever. In reality a cone would always fall down.
Then you can have fun with circuits that generate a billion giga volts or consume a billion giga amps. And nothing explodes!
Hmm. Thinking that has gotten me into a lot of trouble.
the lamp is barley lighting
Heater! I'm surprised, you've actually come up with a good idea!
I'm now visualizing something like Xbox Kinect (or whatever) where we can grab and plug in virtual components, whoops, wrong way round. Oh no, there's A.I. to finish me off and I can smell smoke, it's the PC, quick get the fire extinguisher! This VR is Very Real indeed!
The sims are an okay way to sanity test, but without the underlying knowledge, you will not have an understanding of what you are doing, to say nothing of what is really possible. You will be shackled to other peoples designs and will likely be unable to proceed any further when the working sim fails in reality. Frustration at best. Play sure, but build the knowledge.
That is great stuff, starts from "Matter" and "Energy" as it should. Definitely Recommended.
And this:
http://www.scribd.com/doc/33152860/Basic-Electronics-Vol-1-US-Navy