Bob Cordell Interview: BJT vs. MOSFET

Jump to Latest
#1,087
No, mosfets are not really BAD, but they are not really GOOD, either.
What we need is a high Gm, high voltage, rugged, power mosfet complement that has very low Crss, Ciss and low Coss. Unfortunately, only tubes have low capacitance, and only vertical powerfets have high Gm. The lateral fets have a fairly low capacitance, but nothing else that is exceptional.
SIT's might be an interesting solution, but they will not directly replace mosfets, and are usually triode like, so they drift with power supply voltage.
 
#1,090
Thank's,
but there are not very different in datasheet between IRF130 and 2SK1058

IRF130 vs 2SK1058
ciss: 650pf vs 600pf
coss: 250pf vs 350pf
crss: 44pf vs 10pf

and which is the change if crss is higher.

PMA excuse me but my english it's very bad and i'm not engineer unfortunately.:bawling: I want understand but it's not easy for me. Thank you! Maxpou🙂
 
#1,092
Folks, I have made all fet power amps, with Hitachi mosfets, and they measured and worked OK. I also made an all fet power amp, with vertical fets, that was to go into commercial production, but I had some problems with failure when the unit was accidently shorted. The only factor that I saw that was really different about IRF140 power fets is that they did not seem to live up to all they were originally promised. At the time, I normally used balanced bridged output stages to make amplifiers of 100W or more. This is because I was worried about second breakdown, but I found, after making a number of power amps with DC power supplies between 30-45V that I could protect these amps with a simple dual DC circuit breaker in the DC power suppliy lines. However, when I tried this same technique with the IRF130 and then the IRF140 series, I found that the output devices tended to destruct before the circuit breaker fired. I also found this to be true with Vfet power devices about 10 years earlier.
Since, I believed the spec sheets for the IRF devices and the sales (information) available at the time, I presumed that I could use the IRF130 or IRF 140 devices at +/- 45V supplies with a single sided push pull output to get 100W operation.
It didn't work then, and I doubt that it would work now, except and unless I used an E-I protection circuit that could instantly detect a short and either open a relay or fire an SCR as a crowbar. That is my experience, and since Bob seems to use only +/- 35 V in his prototype, I would like to see some evidence that he can do higher voltages successfully as well.
Not just a paper design, either. Let's see a production design! You know, one that is given out to 'customers' who are not careful and squawk loudly when their amp blows up and maybe takes out one of their speakers as well. ;-)
 
#1,093
Are the vertical mosfets really resistant against 2nd breakdown? Their thermal dependence of Id at constant Vgs is POSITIVE up to high values, like 10A - 20A. Their Rdson is pretty low, and they are fast (local overheating of the chip??). All this together, compared to negative d Id/dT of laterals (above several hundred mA) and their high Rdson - is not this an answer?
 
#1,095
PMA said:
Are the vertical mosfets really resistant against 2nd breakdown? Their thermal dependence of Id at constant Vgs is POSITIVE up to high values, like 10A - 20A. Their Rdson is pretty low, and they are fast (local overheating of the chip??). All this together, compared to negative d Id/dT of laterals (above several hundred mA) and their high Rdson - is not this an answer?
Click to expand...
Pavel,

There are topologies where these disadvantages are not such.
 
#1,096
john curl said:
Folks, I have made all fet power amps, with Hitachi mosfets, and they measured and worked OK. I also made an all fet power amp, with vertical fets, that was to go into commercial production, but I had some problems with failure when the unit was accidently shorted. The only factor that I saw that was really different about IRF140 power fets is that they did not seem to live up to all they were originally promised. At the time, I normally used balanced bridged output stages to make amplifiers of 100W or more. This is because I was worried about second breakdown, but I found, after making a number of power amps with DC power supplies between 30-45V that I could protect these amps with a simple dual DC circuit breaker in the DC power suppliy lines. However, when I tried this same technique with the IRF130 and then the IRF140 series, I found that the output devices tended to destruct before the circuit breaker fired. I also found this to be true with Vfet power devices about 10 years earlier.
Since, I believed the spec sheets for the IRF devices and the sales (information) available at the time, I presumed that I could use the IRF130 or IRF 140 devices at +/- 45V supplies with a single sided push pull output to get 100W operation.
It didn't work then, and I doubt that it would work now, except and unless I used an E-I protection circuit that could instantly detect a short and either open a relay or fire an SCR as a crowbar. That is my experience, and since Bob seems to use only +/- 35 V in his prototype, I would like to see some evidence that he can do higher voltages successfully as well.
Not just a paper design, either. Let's see a production design! You know, one that is given out to 'customers' who are not careful and squawk loudly when their amp blows up and maybe takes out one of their speakers as well. ;-)
Click to expand...


Thanks for the clarification of your experience, John. It is certainly true that verticals can be less forgiving than laterals or bipolars, and they will go faster than fuses or ordinary circuit breakers. They are fast, and there is very little mechanism inherent in them to limit peak current. This is a fact of life that must be dealt with in verticals, but it is certainly not insurmountable with good engineering. There are effective, yet non-invasive, short-circuit protection circuits that work very well. There are also driver topologies that are inherently better at protecting the MOSFETs.

The prototype I published was only 50 Watts soley for the reason that it was a demonstration vehicle of the concept. In 1984 I built a 200 wpc version of it that worked very well and survived into shorts. My Athena speaker systems each have four 125-watt MOSFET amplifiers in each enclosure. It would also appear that Halcro has no problem with using verticals in a high-power amplifier.

Cheers,
Bob
 
#1,097
john curl said:
Estuart, being an EE doesn't make you a pro in audio amp design. You have to win some 'races' first. For example, do you think that Chas Hansen isn't aware of the output stage internal loop, first done by Hawksford, and now promoted by Cordell on this website?
He not only knows about it, but he has tried it. However, he told me that it did not make the amp 'sound' any better, so he decided not to employ it.
For us, sound quality is the sole criterion for real success.
Click to expand...

Hi John.

You doesn't answer my question, instead you are questioning my reputation as pro in audio amp design. Admittedly, an EE title alone doesn't make you a pro. For example, in our country, carbon speaker cable are sold by an EE :rofl: A fraudulent and abject practice.

As for Chas Hansen, I don't care about his opinion about EC. Maybe his ears weren't accustomed to a distortion free amp, or even worse, he wasn't able to implement Hawksford's idea correctly. I don't know, do you?

As for "For us, sound quality is the sole criterion......"
Who are these "us" ? People who like tonal coloration of an amp?
I bet that the majority of this forum opt for minimal distortion.

But first, please answer my question @ post#1072. For ease of convenience I repeat it her:

If I understand you well, local FB is OK, but applied to more than one stage it's suddenly a crutch? I find this a rather curious theorem that needs to be substantiated with scientific (not sonic!) arguments.
And why you make an exception for Halcro? Is Bruce Candy the one and only guy on earth who is allowed to use global FB?
 
#1,098
estuart said:
Huh? 512w rms, pure class-A, and only 350W idle? Not >1024W idle?
Click to expand...

G'day Edmond.

The two (bridged) class A output stages used per channel idle at 8A, each with a seperate powersuppply of +/-10V. The 'grounds' of each supply are floating, each being driven by a class A/B stage, so that they track the output voltage. It's an old Technics idea that really never took off, marketing wise.


Indeed, an AP is rather expensive. I solved this problem in a different way, by means of a professional sound card, a bank of passive filters and a S/W program (written by myself) to subtract the fundamental and distortion produced by sound card itself. Moreover. This progarm averages the residual over million cycles or so. As result, I can see the distortion products, that would otherwise completely immerse in noise, down to a level of -140dB.
Click to expand...


Yup, something along those lines is pretty much what I'm now looking at. I almost fell over when I saw what even the old "System One" AP's are selling for 😱


Sure, in isolation, a class-A O/P stage is always better. But if one can build a class-B amp with 1ppm distortion I don't bother to bring it further down to say 0.3ppm using class-A.
Click to expand...


No sense of adventure, that's all I'm going to say. :tongue:


"I noticed that you do spice simulations by request" :nownow:
But for you I'll make an exception, although it took me 3 hours to capture your schematic.
Hopefully, you don't mind that I'have changed your design slightly, because I don't have models of MJL19114/4 and I have R28/29 given a value of 47R and C5/6 150pF, as they were unspecified (see attachment). These values give a unity gain frequency of the global FB loop of 1.2MHz. Was that your intention?
Click to expand...


Yes, I know. :sorry: It certainly was a little bit cheeky of me to ask, considering the amount of work involved. Thank's a lot though, I appreciate it. Your loop compensation is pretty much what I was expecting. I havent specified component values yet, as I like to optimise my compensation on the real prototype.

The results, for such a simple design are pretty good, I reckon. I hear the point you have demonstrated about TMC. The advantage of class A MC over class B TMC is obviously marginal. There is a limit to how much I can agree with you though, as mikes might be listening. I don't want to make him angry 🙄

I'm still going to stick with class A, MC. My amp will double as a foot warmer in winter. I was almost persuaded to use class A TMC, but I've already layed out the board. I don't think my ears are good enough to pick 10ppm less or so :spin:

Cheers,
Glen
 
You must log in or register to reply here.
Top Bottom