Yes, adding a cap at the right place can allow higher dissipation for a short term. A cap can function as memory element here of course.
This is not very accurate, but it is a simple measure that can give a bit more useable output under complex drive conditions.
As for the temperature compensation, if you read D Self book you see how difficult it is to sense the temp of the transistor die, and that is what matters. I think the best we can do is compensate for the average temp and err on the safe side, and the heatsink temp is adequate for that.
Jan Diddden
This is not very accurate, but it is a simple measure that can give a bit more useable output under complex drive conditions.
As for the temperature compensation, if you read D Self book you see how difficult it is to sense the temp of the transistor die, and that is what matters. I think the best we can do is compensate for the average temp and err on the safe side, and the heatsink temp is adequate for that.
Jan Diddden
janneman said:
Hi Jan,
Muting is only done when DEAD SHORT is encountered and results in no pop/Clicks....
Input Signal is attenuated in a proportion required only when limiting is needed in presence of reactive loads to safe guard the amp....
Mikeks uses 2 BJT's + passive components..OK low cost excellent solution,
I use 1 Quad Comparator IC[For sensing V & I] + 1 VCA IC[For limiting & muting] + passive components...not costing a fortune to implement...So from where does the BOX full of components arrived in your mind....Try to BE GENEROUS
BTW what are the requisite dimensions of your BOX..L X W X H
AndrewT said:
Hi AndrewT,
Have you ever compared the NVMOS with a similar rating BJT......PRACTICALLY, If you had, you would never say such things in this way....
What makes you to think that THE SHORT TERM CAPABILITY of VERTICAL MOSFETS is JUST the SAME of BJT's....
Why not try some EXploration on the internet regarding these issues.....at WWW.ADVANCEDPOWER.COM, WWW.IRF.COM
The Mosfet Manufacturers...Read some of their application notes regarding the Superiority of Mosfet over the BJT's
and this one too from a manufacturer of Vertical Mosfet Proamps...
http://www.ecler.com/enter/product/info2.htm
http://www.ecler.com/enter/product/SPMNAV.htm
regards,
K a n w a r
Hi Janneman,
just to confirm my statement on Temp compensation.
This thread is about output protection and I was referring to the VI limiter possibly benefiting from some kind of Temp comp.
In other words a VI limiter that allows maximum current draw when the output devices are cold and reduces the permitted current as the output stage warms up.
It is easy to fix this at one selected and typical temperature that the designer believes to be representative of either the pro or domestic environment.
For this single temperature setting the VI limiter will operate too early if the output stage is cooler and allow excessive dissipation if too hot. The too cold limit then becomes audible without need and the too hot situation reduces reliability.
This is different from and additional to the Temp comp that must be fitted (BJTs & VFETs and maybe others in ClassAB) to control the output bias conditions.
just to confirm my statement on Temp compensation.
This thread is about output protection and I was referring to the VI limiter possibly benefiting from some kind of Temp comp.
In other words a VI limiter that allows maximum current draw when the output devices are cold and reduces the permitted current as the output stage warms up.
It is easy to fix this at one selected and typical temperature that the designer believes to be representative of either the pro or domestic environment.
For this single temperature setting the VI limiter will operate too early if the output stage is cooler and allow excessive dissipation if too hot. The too cold limit then becomes audible without need and the too hot situation reduces reliability.
This is different from and additional to the Temp comp that must be fitted (BJTs & VFETs and maybe others in ClassAB) to control the output bias conditions.
Hi Workhorse,
I am either unable to understand your argument or just disagreeing with you. I will need to do more research to decide into which camp I fall.
I have compared your stance to Ecler and there appear to be many similarities, but no facts pertaining to FET vs BJT short term output dissipation. Many unsubstantiated statements though.
Ecler say
One other item intrigued me:-- Gain bandwidth product. Their comparison seemed out of date with regard to modern components or just plain wrong. Can you clarify where your amps/design philosophy lie on this bandwidth line?
I cannot access advanced power. Is my browser set up wrongly?
However, none of this is really what this thread is about. So from your point of view, is it possible to meet my best case senario?
To save you looking back, a design and instructions to implement in our own amplifiers.
I am either unable to understand your argument or just disagreeing with you. I will need to do more research to decide into which camp I fall.
I have compared your stance to Ecler and there appear to be many similarities, but no facts pertaining to FET vs BJT short term output dissipation. Many unsubstantiated statements though.
Ecler say
One other item intrigued me:-- Gain bandwidth product. Their comparison seemed out of date with regard to modern components or just plain wrong. Can you clarify where your amps/design philosophy lie on this bandwidth line?
I cannot access advanced power. Is my browser set up wrongly?
However, none of this is really what this thread is about. So from your point of view, is it possible to meet my best case senario?
To save you looking back, a design and instructions to implement in our own amplifiers.
AndrewT said:
Understood, that was also my take on it, to optimise available power depending on temp. Probably I didn't put it clear enough.
Jan Didden
AndrewT said:
Hi AndrewT,
Why not do some more research in practical way:
Take your Highest rated MODERN POWER BJT [From ONSEMI or TOSHIBA]of Present day which you seem Ultra Superior and then compare it with Same rating N-Channel Vertical MOSFET......
And see the results your self....
And Now If you have the Access to www.Advancedpower.com, you should have sighted the Truth....
We have also tested the new BJT's MJL4281 350V 16A 230W from the ONSEMI which were very superior from their previous counterparts..very much robust are these devices are, but again these also failed when SOA exceeded even marginally due to second breakdown phenomena, Whereas Mosfet APT30M85BVR 300V 40A 300W survived this test........
K a n w a r
Hi Workhorse,
I intend doing a bit more to find what, if any, advantages the lateral and vertical FETs can offer over their BJT brethren.
I will not take your lead and only research
The only reason for using VFETs in the output stage seems to be to save money, deliver very high power, at commercial gear quality, at a competitive price and earn a profit for the business. A skewed conclusion? Probably, but based on insufficient evidence that I have found so far, much of it I suspect being based on unfounded statements made by others. Gullible? or still willing to learn?
I intend doing a bit more to find what, if any, advantages the lateral and vertical FETs can offer over their BJT brethren.
I will not take your lead and only research
, else that may draw me down the same route as you and may deliver a skewed conclusion.
The only reason for using VFETs in the output stage seems to be to save money, deliver very high power, at commercial gear quality, at a competitive price and earn a profit for the business. A skewed conclusion? Probably, but based on insufficient evidence that I have found so far, much of it I suspect being based on unfounded statements made by others. Gullible? or still willing to learn?
Hi,
just pulled up data on APL602L (To264).
5.5A at 100V (550W) at DC and 35A at 100V for 1mS . Very good numbers for a single device. Improvement from DC to 1mS 6.3times.
Input capacitance:- 7.48nF at 10V, rising to 13nF at 1V. cf. lateral 0.6nF and BJT 0.3nF, but both are at lower power.
This type of VFET will never be attractive to me for any of my power amps.
Workhorse,
can you suggest a better alternative, preferably complementary pairs, for me to consider?
just pulled up data on APL602L (To264).
5.5A at 100V (550W) at DC and 35A at 100V for 1mS . Very good numbers for a single device. Improvement from DC to 1mS 6.3times.
Input capacitance:- 7.48nF at 10V, rising to 13nF at 1V. cf. lateral 0.6nF and BJT 0.3nF, but both are at lower power.
This type of VFET will never be attractive to me for any of my power amps.
Workhorse,
can you suggest a better alternative, preferably complementary pairs, for me to consider?
AndrewT said:
Hi AndrewT,
HA HA HAAAAAA.......😀 😉
Why you have choosen this MOSFET only...TRY APT30M85BVR
And check out its Specs.....
Also IRFP260 from IRF compare it with MJL21196 ..see the difference......
The Gate Capacitance of All VFET are high , But the Main thing of Concern is GATE CHARGE Qg which is very low in APT & IXYS Mosfets...
Secondly My Views are only Meant For VERTICAL Mosfets only...Not Lateral ones....
High Power Vertical Mosfets are N-channel ones only ..P-channel versions are of moderate power....IRFP9240, IXYS24N50
Secondly all the INDUSTRIAL LEVEL POWER CONVERSION IS Based on N-channel Mosfets/IGBT's only....which is why these are easily available at low cost ...
Have you considered one thing, "Why Industrial Grade Power Solutions are often built with Mosfets...Not with the BJT's"
because Switching speed of Mosfets is very very high than Bipolars....and also due to absence of SecondBreakdownVoltage
Its your Lack of Understanding about the Vertical Mosfets...If these facts which I quoted were based on UNFOUNDED Statements , then the whole Industrial Power Conversion Solutions Manufacturers were on their way to SHUTDOWN, because you cant make something which is based unfactual information.....
regards,
K a n w a r
Hi Workhorse,
had another look at Advanced. This time APT30M85BVR.
3A @ 100Vdc and 45A @100V 1mS. Very impressive.
Compared to Lateral where the 10mS factor is about 3times and IRFP240 is 6times at 1mS.
The BJTs fare a lot better for short term load factor increase.
The 1S to 10mS factor for 4281 is 5times and for A1943 Dc to 1mS is 48times.
So my conclusion is that lateral and some verticals cannot compare in short term load increase with BJTs. However the APT30M85BVR you have selected has a very wide spread and matches the exceptional A1943 because the A1943 has such a poor high voltage DC SOAR. In this case you are right, APT30M85BVR has a good resistance to short term loading.
When you refer to APT30M85BVR having good resistance to SOAR excursion, do you mean outside the 1mS or 10mS as appropriate for the single shot test signal or do you mean outside the DC SOAR? I would like to know if you are comparing BJTs to APT30M85BVR within the manufacturer's spec limits?
But, I come back to input capacitance 4nF to 5.5nF is still too high to be an easy load for the drivers at high frequency.
Can you explain how to use the Cq data you referred me to? How do you turn "charge" into an equivalent load for the driver, at any frequency?
Finally, I looked up Advanced and selected "Discrete - linear transistor (mosfets) - T0264" (I did not recognise the other case styles) and then compared data. I was not trying to be difficult.
had another look at Advanced. This time APT30M85BVR.
3A @ 100Vdc and 45A @100V 1mS. Very impressive.
Compared to Lateral where the 10mS factor is about 3times and IRFP240 is 6times at 1mS.
The BJTs fare a lot better for short term load factor increase.
The 1S to 10mS factor for 4281 is 5times and for A1943 Dc to 1mS is 48times.
So my conclusion is that lateral and some verticals cannot compare in short term load increase with BJTs. However the APT30M85BVR you have selected has a very wide spread and matches the exceptional A1943 because the A1943 has such a poor high voltage DC SOAR. In this case you are right, APT30M85BVR has a good resistance to short term loading.
When you refer to APT30M85BVR having good resistance to SOAR excursion, do you mean outside the 1mS or 10mS as appropriate for the single shot test signal or do you mean outside the DC SOAR? I would like to know if you are comparing BJTs to APT30M85BVR within the manufacturer's spec limits?
But, I come back to input capacitance 4nF to 5.5nF is still too high to be an easy load for the drivers at high frequency.
Can you explain how to use the Cq data you referred me to? How do you turn "charge" into an equivalent load for the driver, at any frequency?
Finally, I looked up Advanced and selected "Discrete - linear transistor (mosfets) - T0264" (I did not recognise the other case styles) and then compared data. I was not trying to be difficult.
AndrewT said:
Hi AndrewT,
APT30M85BVR is 3A @ 100V @DC & 50A @ 100V@1mS , correct it....
Lets Compare 100mS
MJL4281 is 1.89A @ 100V @ 100mS
APT30M85BVR. is 5A @ 100V @ 100mS
Clearly Doubles the SOA is Mosfet APT30M85BVR.
Now, 10mS
MJL4281 is 5A @ 100V @ 10mS
APT30M85BVR. is 15A @ 100V @ 10mS
3 times more SOA is APT30M85BVR.
So from this Mosfet Clearly Wins the race....
Ok now the SOA Excursion....
I have tested them them OUTSIDE the LIMITS Specified by the manufacturer.....whether DC or Pulse 10mS-100mS
MJL4281 Fails instantly Due to Second Breakdown Voltage
APT30M85BVR. Fails after some finite Time ranging from 15 to 25 % overration...
Now Qg the Gate Charge...
Its relation ship is as follows..
Ig[Gate current for charging the mosfet in Amperes] = Qg[Gate charge in nC] X Frequency of operation in hertz
regards,
K a n w a r
Hi AndrewT,
Here's another manufacturer link,..
http://www.mtxaudio.com/caraudio/technology/ampNChannel.cfm
K a n w a r
Here's another manufacturer link,..
http://www.mtxaudio.com/caraudio/technology/ampNChannel.cfm
K a n w a r
Hi Kanwar,
I think MTX missed the point entirely. The P channel devices are not too slow in the audio bandwidth. As you know, compensation is possible.
-Chris
I think MTX missed the point entirely. The P channel devices are not too slow in the audio bandwidth. As you know, compensation is possible.
-Chris
Hi amp man 1
Can’t find any patents searching at European Patent Office and US patent office
These are u r mentioned manufacturer patents on pure N-channel amps
Smart Engage™ Auto-Turn-On (Patent No: US06556683)
Patented Pure N-Channel Technology (Patent No: US05631608, US05783970)
Patented Power Supply (US05598325)
I want to know about your amps and cast in India
U R website look nice
Cheers,
ravi
Can’t find any patents searching at European Patent Office and US patent office
These are u r mentioned manufacturer patents on pure N-channel amps
Smart Engage™ Auto-Turn-On (Patent No: US06556683)
Patented Pure N-Channel Technology (Patent No: US05631608, US05783970)
Patented Power Supply (US05598325)
I want to know about your amps and cast in India
U R website look nice
Cheers,
ravi
- Status
- Not open for further replies.