I really like 2SC2837 and 2SA1186 in TO-3P. I just cannot find the Spice model for them at all. Apparently quite a few people looking for the model in the pass few years here already and no reply. Sanken site has nothing. If anyone have the model, please help me.
Other than that, Any suggestion on another driver transistor that has LTSpice model available? I am driving 5 power BJT, so it has to be on the hefty side.
Thanks
Other than that, Any suggestion on another driver transistor that has LTSpice model available? I am driving 5 power BJT, so it has to be on the hefty side.
Thanks
Bob Cordell's book tells you how to extract SPICE model parameters, yourself. It might be another possibility.
I am still reading Cordell's book on that. But the data sheet doesn't have enough info. Particular no info on Ccb and Cbe.
I do not have those Sanken models either, but try to use something similar to get yourself going.
Finish reading the book and concentrate on the spice section, it is well written. It isnstructs you to Go to the Cordell web site and get his models.
Search, I have placed my bjt & jfet models, user "rsavas" in the thread for Bob's book. I have accumulated and includes Bob's models stuff I got from mfg's. Stick with Bob's models as a start for sure.
http://www.diyaudio.com/forums/soli...ifferential-pa-simulation-possible-build.html
Also read through this long thread, lots of great discussions about parts of the book
http://www.diyaudio.com/forums/solid-state/171159-bob-cordells-power-amplifier-book.html
Finish reading the book and concentrate on the spice section, it is well written. It isnstructs you to Go to the Cordell web site and get his models.
Search, I have placed my bjt & jfet models, user "rsavas" in the thread for Bob's book. I have accumulated and includes Bob's models stuff I got from mfg's. Stick with Bob's models as a start for sure.
http://www.diyaudio.com/forums/soli...ifferential-pa-simulation-possible-build.html
Also read through this long thread, lots of great discussions about parts of the book
http://www.diyaudio.com/forums/solid-state/171159-bob-cordells-power-amplifier-book.html
Last edited:
What has worked fairly well for me is to buy a dozen of the real life parts, take actual current-versus-voltage measurements on them, and then fit the parameters of a SPICE model to my real-life measurements, following Cordell's instructions. Just in case, you know, the datasheet happens to be several revisions older than the current manufacturing process.
Last edited:
A tutorial on spice model construction is always helpful.
So if I take these device measurements then how do I apply them to the spice model parameters. Looks like I am going to have re-read chapter 20 in Bob's Audio PA book, just to get a vague understanding.
If the datasheets are potentially out of date, what does that say for the spice models, sounds like they are even in worse shape.
So if I take these device measurements then how do I apply them to the spice model parameters. Looks like I am going to have re-read chapter 20 in Bob's Audio PA book, just to get a vague understanding.
If the datasheets are potentially out of date, what does that say for the spice models, sounds like they are even in worse shape.
US President R. Reagan loved to quote a Russian proverb, "Trust, but verify."
It incorporates a certain mindset; a philosophical outlook which acknowledges the possibility of disagreement between (a) the datasheet; (b) the SPICE model; (c) measured data on real parts. Many people refuse to entertain the idea that this might be even slightly possible.
It incorporates a certain mindset; a philosophical outlook which acknowledges the possibility of disagreement between (a) the datasheet; (b) the SPICE model; (c) measured data on real parts. Many people refuse to entertain the idea that this might be even slightly possible.
I guess I am with Reagan or the Russians on this one.
Design without verification is risky business ;-) I guess we call it testing. NASA does it all the time, don't they? Kinda hard to simulate deep space exploration. It's all in the models, eek
Design without verification is risky business ;-) I guess we call it testing. NASA does it all the time, don't they? Kinda hard to simulate deep space exploration. It's all in the models, eek
You should of just asked me. 🙄
2sa1837/2sc4793 models will simulate those big to-3p's nearly perfectly.
The toshiba's just have slightly faster Ft , Hfe - saturation - Vbe/thermal is
REAL close.
OS
You quote the right President. Do I miss President Reagan.
I still need to read Cordell book, I just started on chapter 20, still need to learn how to test the transistors for parameters.
Fairchild fooled me!!! When I looked for the model files of KSA1381 and KSC3503, Fairchild made me sign up with them. Then after that, I didn't see any of the model.
Last edited:
.MODEL 2SA1837 PNP ( IS=2.39372559E-10 NF=1.304015937 BF=300 VAF=273 IKF=2.087725944 NK=0.94719458 ISE=1.46829699E-11 NE=1.526663542 BR=4 NR=1 VAR=20 IKR=1.05 RE=0 RB=1.8 RC=1.65 CJE=4.7407E-10 VJE=1.1 MJE=0.5 CJC=8.6700E-11 VJC=0.3 MJC=0.3 TF=1.642191E-09 FC=0.5 ITF=1.076260106 XTF=5.868994022 TR=1.38U)
.MODEL 2SC4793 NPN ( IS=1.8E-09 NF=1.43 BF=146.38 VAF=273 IKF=2.6 NK=0.95 ISE=6.286997E-10 NE=2.223629 BR=4 NR=1 VAR=20 IKR=1.05 RE=0 RB=1.7 RC=1.25 CJE=5.96964E-10 VJE=1.1 MJE=0.5 CJC=5.78E-11 VJC=0.3 MJC=0.3 TF=1.22678E-09 FC=0.5 ITF=10 XTF=99.52253015 TR=983N)
.MODEL 2SC4793 NPN ( IS=1.8E-09 NF=1.43 BF=146.38 VAF=273 IKF=2.6 NK=0.95 ISE=6.286997E-10 NE=2.223629 BR=4 NR=1 VAR=20 IKR=1.05 RE=0 RB=1.7 RC=1.25 CJE=5.96964E-10 VJE=1.1 MJE=0.5 CJC=5.78E-11 VJC=0.3 MJC=0.3 TF=1.22678E-09 FC=0.5 ITF=10 XTF=99.52253015 TR=983N)
Thanks, too bad these two are a little under power at 1A max. I'll be happy to use them if they are a little more stronger.
don't just read the headline specs.
LOOK at the SOA plot.
The 1837 is 12W DC @ 40Vce. That is what you should be comparing for selecting or rejecting devices for YOUR supply rail voltage.
Note that the 12W (0.3A @ 40Vce) is good for beta droop on both the 25°C & 125°C plots.
Cob is a bit high.
Shame we can't get guaranteed genuine full specification Hitachi 2sb649/d669.
I have not found any device that can better those.
I have bought clones as well as Hitachi versions but am unsure about how close they are in spec. I even imported Hitachi from the west coast of the USA. What a pile of rubbish they sent me.
Last edited:
I read a few more pages of chapter 20.3 of Cordell's book. Looks like it's very well written with detail explanation on how the parameters works and how to get them.
Problem with me is I don't have the different power supplies. I got a lot more models from Linear Technology, I got the models of MJL and KSA and KSC devices.
How's the models from Linear technology? Also I am thinking about using their model to simulate just like what Cordell described in chapter 20.3. Then I compare to the data sheet to see how they match up. Then massage the model to fit the data sheet. Of cause this is not going to be as good as measuring the real device, but this is a big step in verifying the model.
What do you think?
Thanks
BTW. I am living in Sunnyvale, we might be living close by. If you are interested, we can meet up some day. I am new in audiophile, but I have been an engineer for almost 30 years and retired.
Ha ha, I am not after your dual JFET!!! I only buy stuffs in production. I already have my eye on Interfet U232 for $13 from Mouser.
Last edited:
Thanks a million!!
I spent a lot of time last night going through the models in your txt files. I found most of the transistors I am interested. That is very helpful.
Question is how accurate are their models? As Mark Johmson talked about verifying the model and massage the parameters. I don't have power supplies to do on the real device. I am thinking about using the model and simulate like Cordell's book, then compare to their data sheet to massage the model so the model fits the data sheet.
In this case, IF the transistor agree with the data sheet, then the model should be accurate enough.
Thanks
okay then try from this selection.
Transistors Bipolar - BJT | Mouser
I have no idea, as I am not smart enough to figure out their quality. I would however put more trust in Bob Cordells' models. So this is what I do, I use his models to start with, then try changing with other models to see what differences they make, one or two at a time.
I know there is a problem with the ksa1220/c2690 models that I have there, they hang the simulation.
Did I get them from OS 🙂 I really do not remember
Last edited:
I figure Bob's models are more accurate as he did what he outlined. That make me want to go back to the more conservative route and use the ones from Bob's model......MJL3281, MJL1302 for power BJT. MJE15032 MJE15033 for driver.
I still like the KSC3503 and KSA1381. Just wonder how accurate are the Linear Technology on this particular two.
Thanks
- Status
- Not open for further replies.