Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
- LTspice
- Messages
Search
|
Re: Ltc3765 & 3766 combo dc/dc simulation goes wild
Tony Casey
--- In LTspice@..., "oleka111" <oleka111@...> wrote:
I have received a response from Linear (not Mike - I think he must be on holiday). Basically, what they say is that the behaviour is as expected, and related to the undervoltage lockout on Vcc, which is connected to the regulated output in this circuit, and that this is operating the device somehow incorrectly because of that. However, the LTspice simulation is almost exactly the same as the application circuit in the datasheet (with the exception that the output is not floating), and the simulation would suggest it won't work as published with 5% resistors. The datasheet says: the minimum Vcc voltage is 5V, and that the regulated DC voltage must be above this if it supplies Vcc directly, or an overvoltage condition can occur. So, this is kind of a known "feature". With zero tolerance resistors, the calculated regulated output voltage is (1 + 4.42k/604)*599.6mV, which is 4.987V, so this is never going to work in the real world. This is what the LT response actually said: "Alright the jig circuit shows that you can use 4.42KOhm, and it is not suppose to take that as the real design. If you want the design close to the scenario, you need to go through from more realistic perspective. Anyway, the model is right, and the jig is right. If you argue that the circuit is not a robust design, that is not the issue of the model. Many times our jig circuits show simplified setup, for example, more ideal magnetic devices, no ESL of sensing resistor, down sized soft-start cap, etc. It is there for user to quickly pick up the fundamental features of the part, and we encourage users to design their applications based on that." For what it's worth, I modified the schematic so that the Vcc pin was fed from a separate 5V supply and reduced the feedback resistor to 4k1 (from 4k42), and the output did regulate properly at 4.67V. Hope that helps somebody. Regards, Tony |
|
Re: Ltc3765 & 3766 combo dc/dc simulation goes wild
With 100Ohm load I got these results (rval is the value of R11 in the
toggle quoted message
Show quoted text
scheme): from error log with .MEAS TRAN res1 FIND V(out) AT=0.88m .step rval=4300 .step rval=4200 .step rval=3900 Measurement: res1 step v(out) at 1 4.99022 0.00088 2 4.88817 0.00088 3 4.7575 0.00088 4 6.25114 0.00088 It is the same behavior as with 1KOhm load but the critical value of R11 is lower. Regards. Varoli Il 19/07/2013 11.55, Helmut ha scritto:
--
Prof. Vincenzo Varoli Politecnico di Milano Dip. Energia Via G. Ponzio 34/3 I20133 Milano Italia Tel. 0223996393 FAX 0223996309 |
|
Re: Ltc3765 & 3766 combo dc/dc simulation goes wild
Hello Olek,
toggle quoted message
Show quoted text
Have you tried with more load, e.g. 100Ohm instead of 1kOhm? Best regards, Helmut --- In LTspice@..., "oleka111" <oleka111@...> wrote:
|
|
Re: Ltc3765 & 3766 combo dc/dc simulation goes wild
Nice to see that you spotted it too......
toggle quoted message
Show quoted text
What if real circuit would behave the same way ? Hope Linear will fix this issue soon. I didn't post the original file, because everyone can load it as an LT's example named 3765.asc and see the effect of diminishing of the voltage feedback resistor. Regards, Olek A. --- In LTspice@..., "Tony Casey" <tony@...> wrote:
|
|
Re: UAF42
That one is easy. Its "just" a couple of op-amps. follow the block diagram in the spec sheet and you pretty much have it.
toggle quoted message
Show quoted text
Jim Wagner Oregon Research Electronics. On Jul 18, 2013, at 3:18 PM, Suleiman wrote:
Hello. |
|
Re: LTspice Genealogy - The Heritage of Simulation Ubiquity
Hello, a.s.
toggle quoted message
Show quoted text
I'm impressed by the work you have put in on this. I would like to add one note, which I have peripherally mentioned in the list before. Newcomers are likely to not remember or never have known what the "state of computing" was when spice began in the early 1970s. I encountered it on machines limited to punch-card input and line printer output, only. You submitted "job decks" (and some still refer to a netlist as a "spice deck"); woe unto you who dropped one of those punch card boxes! And the reams of fanfold paper needed to print out the initial node conditions plus the response of selected nodes. The response was an "ASCII graph" with an even time step. I don't know if it was computed on an even time step or if it was interpolated after the fact. An important part of this as that as long as the machine ran FORTRAN, accepted punch card input and did line printer output, spice would run. There were no issues of "operating system" or cross-platform behavior. I saw it run on big IBM mainframes and CDC6400s. I think that it was the graphic user interface that really pushed various implementations into one OS or another. It was simply too hard to do a two-OS implementation (for the most part). So, maybe one of the important anchor points in your timeline ought to be when GUIs began to be made as an integral part of the various spice versions. Thans for all your great work. Jim Wagner Oregon Research Electronics On Jul 18, 2013, at 9:08 AM, analogspiceman wrote:
Please visit this new page at the LTwiki. |
|
Re: inductance with a permeability in dependency of frequency
--- In LTspice@..., "sawreyrw" <sawreyrw@...> wrote:
My mistake. Here's the web page and correct link: Other related stuff there. RL |
|
Re: looking for a model for AD8363
Tony Casey
--- In LTspice@..., "gift4jo" <gift4jo@...> wrote:
Perhaps if you had included all of your requirements in your post, instead of just "...any other RMS-DC converter", you might have avoided needlessly further increasing the entropy of the universe. If you really want an RF RMS to DC converter, you mighty want to check out ADI: Maxim do them too: You won't find any models there, but I have created behavioural models from the datasheets, but unfortunately I can't share them with you as someone else paid for it. But it's not that hard. Don't expect the same accuracy as you would get from a proper LF RMS to DC converter, though. Regards, Tony |
|
Re: LTspice Genealogy - The Heritage of Simulation Ubiquity
--- In LTspice@..., "analogspiceman" <analogspiceman@...> wrote:
Hello analogspiceman, Thanks a lot for collecting all the mile stones. I always wonder that 1999 was the "birth" of LTspice. I personally discovered that LTspice is a SPICE program in the summer 2001. I still believe that I was one of the early adopters using it as a general SPICE program. Also the discussions about LTspice started in 2001 in the usenet, sci.electronics.design and sci.electronics.cad. Best regards, Helmut |
|
Re: Ltc3765 & 3766 combo dc/dc simulation goes wild
Tony Casey
--- In LTspice@..., "oleka111" <oleka111@...> wrote:
Hello, Well spotted. I have tried this distribution example circuit on two machines, and it behaves as you describe. I swept the feedback resistor from 4k1 to 4k3 in 10R steps and found that the LTC3766 transitions from closed loop to open loop behaviour when the value goes below 4210R. Whether this a model error or not, I don't know, but the example circuit is clearly unsatisfactory. I have reported this as a bug to Mike. Regards, Tony |
|
Re: Ltc3765 & 3766 combo dc/dc simulation goes wild
oleka111 wrote:
Running an dc/dc example named 3765.asc with slightly lowered feedback resistor (4.42 --> 4.2k) gives Vout appr. several times bigger thanNobody can answer this because you gave us no information. Try uploading your complete schematic so we can see what is going on. Andy |
|
Re: MPS4250
John Woodgate
In message <1374140838.88699.YahooMailNeo@...>,
dated Thu, 18 Jul 2013, abeill¨¦ jean-claude <jean_claudeabeille@...> writes: I am a french retired computer analyst and a music andIt is quite good. For the moment, I try to understand how does work my amp with LTSpice.Please upload the circuit (we use the US term ''schematic' on this list) to Files => Temp on the list's web site: I asked Bob Cordell some help but he was to busy and answer me to readI expect we can help once we see the schematic. -- OOO - Own Opinions Only. See www.jmwa.demon.co.uk Why is the stapler always empty just when you want it? John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK |
|
Re: MPS4250
abeill¨¦ jean-claude
OK, thank you John.
?Perhaps you can help me in my quest. I am a french retired computer analyst and a music and electonicsenthusiast. Please, will you excuse my english. For the moment, I try to understand how does work my amp with LTSpice. Actually, after an error in handling, one channel failed and activated a protection circuit (too much quiescent on one output stage polarity). I tried to fix this but I failed and ordered a complete replacement PCB. But I am curious and like to understand things. So I drawn the circuit in LTSpice from the PCB and could not understand it in some places. I asked Bob Cordell some help but he was to busy and answer me to read chapter 1-4 from his book. OK. In fact, in LTSpicen there is almost no quiescent current in output stage and the VBE spreader does not work at all. I thought that some models where not correct but I don't know. I'm deadlocked. Jean-Claude. ________________________________ De?: John Woodgate <jmw@...> ??: LTspice@... Envoy¨¦ le : Mercredi 17 juillet 2013 18h55 Objet?: Re: [LTspice] MPS4250 ? In message <ks6d15+3s9u@...>, dated Wed, 17 Jul 2013, jean_claudeabeille <jean_claudeabeille@...> writes: Can anybody here tell me where I can find a spice model for this PNP ?Apart from the packaging, it is similar to a BC556 or BC557. -- OOO - Own Opinions Only. See www.jmwa.demon.co.uk Why is the stapler always empty just when you want it? John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK |
|
Re: Changing the mutual inductance coefficient of K statements with time
John Woodgate
In message <747691374138968@...>, dated Thu, 18 Jul 2013, =?koi8-r?B?4czFy9PBzsTSIOLP0sTPxNnOz9c=?= <BordodunovAlex@...> writes:
My source files without any corrections you simulated?Yes, I deleted the first download and downloaded the zip again, opened it and saved all five files to Desktop. I made no changes. The 'Unknown subckt...' error occurred. I opened Core.lib with UltraEdit and copied the .subckt Winding0 to the schematic as a Spice directive. The simulation then ran. Just to see what would happen, I converted the directive to a comment and the simulation did not run, giving the 'Unknown subckt...' error. I am using Windows XP SP3, but I don't believe this has anything to do with Windows. -- OOO - Own Opinions Only. See www.jmwa.demon.co.uk Why is the stapler always empty just when you want it? John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK |
|
Re: Changing the mutual inductance coefficient of K statements with time
John Woodgate.
toggle quoted message
Show quoted text
My source files without any corrections you simulated? Bordodynov. 18.07.2013, 12:40, "John Woodgate" <jmw@...>: In message <256711374134677@...>, dated Thu, 18 Jul 2013, |
to navigate to use esc to dismiss