Hi Dimitri,
AFAIK, these are proprietary LT parameters. You can't use them with other simulators.
Cheers,
E.
BTW, have a look at Class-I stuff. Very promising: http://www.diyaudio.com/forums/soli...ogy-construction-troubles-40.html#post2789990
AFAIK, these are proprietary LT parameters. You can't use them with other simulators.
Cheers,
E.
BTW, have a look at Class-I stuff. Very promising: http://www.diyaudio.com/forums/soli...ogy-construction-troubles-40.html#post2789990
Hi Dale,
Thanks for your kind words. Much of my inspiration for doing decent models came from Andy Connors.
Cheers,
Bob
Hi Edmond,
These are VDMOS parameters that are proprietary to LTspice. They provide a very good modeling of the capacitance behavior of power MOSFETs without resort to clumsy subcircuits that don't seem to model those capacitances well. As far as I know, there are not any other simulators that do a decent internal modeling of these capacitances, but I could be wrong. Unfortunately, at the same time, the LTspice VDMOS model does not support the EKV model (although I think Mike put something in there that may be a step in the right direction, I'm just not sure). It would be great if LTspice had an EKV-VDMOS model. Of course, I guess this would be proprietary to LTspice as well.
Cheers,
Bob
modeling sub-threshold conductance
Hi Bob,
>It would be great if LTspice had an EKV-VDMOS model.
The bad news is: you can't have them all.
The good news is that LT's VDMOS model also seems to cover the sub-threshold conductance. See: Undocumented LTspice - LTwiki
VDMOS: Breakdown and Sub-threshold Enhancements
LTspice now contains a number of otherwise undocumented parameters to enhance its proprietary VDMOS model. These allow for body diode breakdown and subthreshold conduction with independent fits to the saturation and linear regions of the output characteristics.
BV: breakdown voltage.
IBV: breakdown current at breakdown voltage.
nBV: breakdown emission coefficient.
Mtriode: A conductance multiplier for the triode region. It allows independent matching of the saturation and linear regions of the MOSFET.
subthres: The current (per volt Vds) at which the square-law drain current verses Vgs switches over to exponential. '
I think it's worth to have a closer look at it and pray it switches over without glitches that spoils a THD analysis.
Cheers,
E.
Hi Bob,
>It would be great if LTspice had an EKV-VDMOS model.
The bad news is: you can't have them all.
The good news is that LT's VDMOS model also seems to cover the sub-threshold conductance. See: Undocumented LTspice - LTwiki
VDMOS: Breakdown and Sub-threshold Enhancements
LTspice now contains a number of otherwise undocumented parameters to enhance its proprietary VDMOS model. These allow for body diode breakdown and subthreshold conduction with independent fits to the saturation and linear regions of the output characteristics.
BV: breakdown voltage.
IBV: breakdown current at breakdown voltage.
nBV: breakdown emission coefficient.
Mtriode: A conductance multiplier for the triode region. It allows independent matching of the saturation and linear regions of the MOSFET.
subthres: The current (per volt Vds) at which the square-law drain current verses Vgs switches over to exponential. '
I think it's worth to have a closer look at it and pray it switches over without glitches that spoils a THD analysis.
Cheers,
E.
Hi Edmond,
Good catch! I think this is what I thought I had heard about but did not have details. Will have to try it. Best of both worlds would be wonderful, especially if it is in a good, free, easy-to-use simulator like LTspice.
Cheers,
Bob
Aaahh, you read my mind. I was too busy eating turkey for Thanksgiving.
Maybe I'll get to it this week. I've been having to really bust my butt at work lately. I hate it when work interferes with audio.
This time of year my dear wife also has a honey-do list the length of my arm. She always wants the outside Christmas decorations up right after Thanksgiving - they are not up yet.
Cheers,
Bob
LT1166
Hello,
Does anyone know of an alternative to the LT1166 bias controller Bob mentions in section 27.5 of his book?
This part appears not to be available: I checked the Linear website, and they have 0 dip packages available from the their online store. Lead time is 12 weeks, with a minimum buy of $5000! These chips are not available at Digikey, Mouser, or Allied.
I was thinking that the LT1116 would be a good way to dynamically control bias, as the text suggests. On many DIY transistor based amplifiers I notice that they require manual adjustment via trim pot.
Is the Vbe multiplier a better solution? On page 291 I notice several variations, one including a trim pot in Figure 14.9 (b).... Is this a necessary evil?
/Mason
Hello,
Does anyone know of an alternative to the LT1166 bias controller Bob mentions in section 27.5 of his book?
This part appears not to be available: I checked the Linear website, and they have 0 dip packages available from the their online store. Lead time is 12 weeks, with a minimum buy of $5000! These chips are not available at Digikey, Mouser, or Allied.
I was thinking that the LT1116 would be a good way to dynamically control bias, as the text suggests. On many DIY transistor based amplifiers I notice that they require manual adjustment via trim pot.
Is the Vbe multiplier a better solution? On page 291 I notice several variations, one including a trim pot in Figure 14.9 (b).... Is this a necessary evil?
/Mason
Hi , Edmond
Quite a good idea , and a value of a few kilo ohms would
be even better , about 4.7k should bring reasonnable THD reduction
while not endengering stability conditions too much.
Cheers,
LT1166 Altneratives
Thanks, I'll have to look into the Class-I output topology. I did come across your auto bias page looking for an alternate solution. Interesting stuff!
In regard to the IC solution, I did find a few high speed power MOSFET
drivers, but I'm not sure if they would be a viable alternative to the LT1166
Anyone have an idea of these would work as a dynamic bias spreader?
/Mason
Thanks, I'll have to look into the Class-I output topology. I did come across your auto bias page looking for an alternate solution. Interesting stuff!
In regard to the IC solution, I did find a few high speed power MOSFET
drivers, but I'm not sure if they would be a viable alternative to the LT1166
Anyone have an idea of these would work as a dynamic bias spreader?
/Mason
NO, they will not work as a dynamic bias spreader.
Those are gate drivers for switching ( Class D , SMPS, ...)
Cheers
Stein
There are a number of vendors who seem to have the part in stock
Find LT1166 Stock and Compare Prices Across the Most Reputable Distributors in the Industry.
Hope this helps. I am curious about this part too.
Factory lead time is at least four weeks for the two vendors that carry them: Arrow and Avnet. With Avnet, you have to buy them in batches of 3000.
It's too bad this is such an obscure part... Yet another reason to stick with discrete designs for DIY. One never knows when IC chips will go out of production.
/Mason
TMC transition frequency
Hi Wahab,
Indeed, with R=680 Ohms, C1 = 33p and C2= 180p, the transition frequency was quite high: 1.1MHz. With R = 1k Ft becomes 750kHz, still a little bit high.
OTOH, with 4k7 you get 160kHz. Isn't that too much on the safe side?
As a (crude) rule of thumb I would suggest a Ft of half of the ULGF of the global NFB loop. Making it much lower results in less distortion reduction, which might be considered as 'opportunity loss'. But you are right: stability is much more important than the lowest possible distortion.
Cheers,
E.
Hi Wahab,
Indeed, with R=680 Ohms, C1 = 33p and C2= 180p, the transition frequency was quite high: 1.1MHz. With R = 1k Ft becomes 750kHz, still a little bit high.
OTOH, with 4k7 you get 160kHz. Isn't that too much on the safe side?
As a (crude) rule of thumb I would suggest a Ft of half of the ULGF of the global NFB loop. Making it much lower results in less distortion reduction, which might be considered as 'opportunity loss'. But you are right: stability is much more important than the lowest possible distortion.
Cheers,
E.
Hi Edmond,
Agree with you on the principle , however , after many sims ,
it looks like increasing Ft of TMC network above those 160khz or so
will reduce stability drastically once the load is a little capacitive,
unless the effect is contained by the usual LR output network ,
wich should not be used as a cure to compensate for a marginally
stable design.
Indeed , if the VAS has sufficent gain , 100 to 200KHZ TMC Ft will
be enough to gain almost one order of magnitude in THD reduction,
at least in the audio band , where it matters most.
Cheers
LME49830 + TMC
Hi Homemodder,
I think it's feasible, though I'm afraid that the improvement in distortion reduction will not be that spectacular, as the distortion of the chip itself, around 10ppm, will define the lowest distortion level possible. As for a final answer, well, the proof is in the pudding.
@ Wahab, okay, I understand your considerations.
Cheers,
E.
Hi Homemodder,
I think it's feasible, though I'm afraid that the improvement in distortion reduction will not be that spectacular, as the distortion of the chip itself, around 10ppm, will define the lowest distortion level possible. As for a final answer, well, the proof is in the pudding.
@ Wahab, okay, I understand your considerations.
Cheers,
E.