alant4321 said:
1. better sound?
there is no one better. The ycan sound just as goo BOTH.
If you will go for MOSFET, like me, or BJT like others
is more a matter of taste. Or what power transistor you have at home to use.
2. there is an upper limit how much power you can get out of LME49810.
Because can supply max 60 mA.
now, if all those power transistors to drive (for example for 200 Watt/8 Ohm)
take more base current than 60 mA, then you get into distortion.
Because LME49810 can not give any more current.
Where is the limit, depends on GAIN of Output BJT Devices.
Using higher gain Power Transistors .. you may maybe make 200 Watt.
Nicks said:
I will explore the internet, if someone has posted Spice Model for LME49830.
But LME49830 has not been around for very long.
Maybe is too early to find .....
First search would be at www.national.com
... they have some Spice Model design support .. for download
I post back here, if I find some spice for LME49810 or LME49830
Regards Lineup
Thank you for ¡§lineup¡¨ reply.
1) Depending on individual preference, I personally like MOSFET for output stage in power amplifier because the characteristic of MOSFET is similar to ¡§Tube¡¨ for amplification which feel soft and comfort sound, but it maintains the analytical capability to drive power. It gives more detail sound than ¡§Tube¡¨ as well as it is superior to BJT in feeling of musical sound. Do you concur?
2) In model M-49810-A using 3 pairs of BJT J4315/J4215 as output stage, I have asked CP Design at http://www.chip-pcb.cn for higher output power if increasing the supply voltage, he replied as follow ¡§Voltage and load impedance affect output power. M-49810-A power out 100W(8Ohm/+-45V), power out 300W(8Ohm/+-75V)¡¨. My query is that is the gain of 3 pairs output BJT J4315/J4215 sufficient to give 200W/8ohm output power or more without distortion if increase the supply voltage?
3) LME49810 and LME49830 are mono design while LM4702 is stereo design. If using LM4702B (higher precision), will the sound be compatible to either LME49810 or LME49830?
4) I am looking for DIY kit of either LME49830 or LM4702 using 2 pairs or more MOSFET as output stage per channel. Does anyone have source of it?
1) Depending on individual preference, I personally like MOSFET for output stage in power amplifier because the characteristic of MOSFET is similar to ¡§Tube¡¨ for amplification which feel soft and comfort sound, but it maintains the analytical capability to drive power. It gives more detail sound than ¡§Tube¡¨ as well as it is superior to BJT in feeling of musical sound. Do you concur?
2) In model M-49810-A using 3 pairs of BJT J4315/J4215 as output stage, I have asked CP Design at http://www.chip-pcb.cn for higher output power if increasing the supply voltage, he replied as follow ¡§Voltage and load impedance affect output power. M-49810-A power out 100W(8Ohm/+-45V), power out 300W(8Ohm/+-75V)¡¨. My query is that is the gain of 3 pairs output BJT J4315/J4215 sufficient to give 200W/8ohm output power or more without distortion if increase the supply voltage?
3) LME49810 and LME49830 are mono design while LM4702 is stereo design. If using LM4702B (higher precision), will the sound be compatible to either LME49810 or LME49830?
4) I am looking for DIY kit of either LME49830 or LM4702 using 2 pairs or more MOSFET as output stage per channel. Does anyone have source of it?
I have no idea what gain those power transistors have.
Maybe someone else knows.
Now, There is a possibility that you can maybe use predrivers (like MJE15030/31)
to increase those 60 mA.
If so, then it is only the MAX Voltage of LME49830 that sets upper limit.
About LM4702 and compare to LME49810/830 see my previous post in this topic. Where I give a brief compare.
And my verdict is:
these 2 new chips are much better drivers than those older, like LM4702
My post, with a link!!!!!!!!!!!!!!!!!
http://www.diyaudio.com/forums/showthread.php?postid=1564068#post1564068
Maybe someone else knows.
Now, There is a possibility that you can maybe use predrivers (like MJE15030/31)
to increase those 60 mA.
If so, then it is only the MAX Voltage of LME49830 that sets upper limit.
About LM4702 and compare to LME49810/830 see my previous post in this topic. Where I give a brief compare.
And my verdict is:
these 2 new chips are much better drivers than those older, like LM4702
My post, with a link!!!!!!!!!!!!!!!!!
http://www.diyaudio.com/forums/showthread.php?postid=1564068#post1564068
Yes, correct!
The max. output current are 5.5mA, 60mA and 56mA for LM4702, LME49810 and LME49830 respectively. Since my preference is MOSFET as output stage, then using LME49830 as driver for better current power and with MOSFET as output stage will be my target.
At same output power such as 100W~200W/8-ohm, more pairs of MOSFET => the better sound.
Why no price in M9830-A from http://www.chip-pcb.cn/en/M-9830-A.htm ? Is it ready?
The max. output current are 5.5mA, 60mA and 56mA for LM4702, LME49810 and LME49830 respectively. Since my preference is MOSFET as output stage, then using LME49830 as driver for better current power and with MOSFET as output stage will be my target.
At same output power such as 100W~200W/8-ohm, more pairs of MOSFET => the better sound.
Why no price in M9830-A from http://www.chip-pcb.cn/en/M-9830-A.htm ? Is it ready?
56 mA output current.
How much MOSFET can we drive with this?
I would say my limit would be 2 pairs of IRFP240/IRFP9240.
However, I guess there are MOSFET users,
that would think 50 mA is suitable to drive a lot more.
I have my own rule-of-the-thumb.
There is a value in MOSFET datasheets called Input Capacitance, Ciss
For IRFP240 this is rather low, for be a 150 Watt HEXFET: 'Only' 1300 pF
The drive current I go for is: Between Ciss/100 and Ciss/50 (mA)
Now one pair IRFP240/IRFP9240 has got like 1300+1300 = 2600pF Ciss
A. 2600/100 = 26 mA
B. 2600/50 = 52 mA
If using my rule of thumb lower value of A. 26 mA, I can drive 2 pairs = 52 mA
But as I say, many MOSFET users would think I am a bit restrictive
using so much current per MOSFET.
They would probably try to drive 3 or 4 pairs of IRFP240/IRP9240.
And quite sure this would work
How much MOSFET can we drive with this?
I would say my limit would be 2 pairs of IRFP240/IRFP9240.
However, I guess there are MOSFET users,
that would think 50 mA is suitable to drive a lot more.
I have my own rule-of-the-thumb.
There is a value in MOSFET datasheets called Input Capacitance, Ciss
For IRFP240 this is rather low, for be a 150 Watt HEXFET: 'Only' 1300 pF
The drive current I go for is: Between Ciss/100 and Ciss/50 (mA)
Now one pair IRFP240/IRFP9240 has got like 1300+1300 = 2600pF Ciss
A. 2600/100 = 26 mA
B. 2600/50 = 52 mA
If using my rule of thumb lower value of A. 26 mA, I can drive 2 pairs = 52 mA
But as I say, many MOSFET users would think I am a bit restrictive
using so much current per MOSFET.
They would probably try to drive 3 or 4 pairs of IRFP240/IRP9240.
And quite sure this would work
Hi,
56Vpk into a resistive load of 8r0 requires 7Apk from the BJT output stage.
But the requirement to feed a reactive load (a speaker) is much more onerous.
Expect to have to supply at least double this peak current and for the most severe speaker loads approaching three times the resistance current on short starting/stopping transients.
To get ~20Apk from the BJT output stage will require a drive current of at least 500mA. To this must be added the drive to charge and discharge the output stage capacitance. Allow maybe another 20mApk.
There is no way the chip driver stage specified @ 60 or 56mA can ever get close to driving a 200W output stage connected to 8ohm speakers without an intervening driver stage.
56Vpk into a resistive load of 8r0 requires 7Apk from the BJT output stage.
But the requirement to feed a reactive load (a speaker) is much more onerous.
Expect to have to supply at least double this peak current and for the most severe speaker loads approaching three times the resistance current on short starting/stopping transients.
To get ~20Apk from the BJT output stage will require a drive current of at least 500mA. To this must be added the drive to charge and discharge the output stage capacitance. Allow maybe another 20mApk.
There is no way the chip driver stage specified @ 60 or 56mA can ever get close to driving a 200W output stage connected to 8ohm speakers without an intervening driver stage.
For you guys asking about using BJT output stage, take Andrew's advice and design an extra current gain stage into the design. The LME49830 boards from National using 2 devices per side (Toshiba FETs, you can see the info in AN-1850) is fine with no extra current gain stage. But if you go for BJT's then you need more current drive. You can always short out the extra current gain stage if you decide it is not needed and the added PCB space is pretty minimal. But I think you will see that it is needed for the power ranges mentioned.
-SL
-SL
For your Reference: Using HEXFETs in High Fidelity Audio at http://sound.westhost.com/articles/hexfet.htm
HEXFET Power MOSFET as output stage: (Application in M-9830-A with LME-49830TB as core driver)
IRFP240 data sheet at http://www.irf.com/product-info/datasheets/data/irfp240.pdf
IRFP9240 data sheet at http://www.irf.com/product-info/datasheets/data/irfp240.pdf
HEXFET Power MOSFET as output stage: (Application in M-9830-A with LME-49830TB as core driver)
IRFP240 data sheet at http://www.irf.com/product-info/datasheets/data/irfp240.pdf
IRFP9240 data sheet at http://www.irf.com/product-info/datasheets/data/irfp240.pdf
SpittinLLama said:
Yes, agree SpittinLLama
Especially whe using 2SK1530/2SJ201
or 2SK1058/2SJ62
we can use LME49830 without extra predrivers.
These Toshibas are more Lateral Devices and more happy with 56 mA drive.
That is, we can not even connect too many Lateral pairs on output.
But say 2-3 pairs in parallel (=4/6 MOSFETs) would be okay, in my opinion.
This would still be able to perform very high class audio.
mlloyd1 said:
there is no language problem, unless you can not read my english
my english strives to be very basic - considering the many chinese persons reading my posts
I get some 10 personal emails daily .. from Asia!
Would be a big problem, mostly for you, if we talked swedish, my language
simple math has got no language
more probably my math is wrong
But if you say the output of LME49830
sees no more than 1300 pF from one IRFP240/IRP9240 pair
no one is more delighted, than me, Lineup
Great!
thanks mlloyd1
As per National Semiconductor Application note AN-1850 at http://www.national.com/an/AN/AN-1850.pdf , recap some statements as below:
1) The LME49830 IC in theory can provide output power levels in excess 1 kW in combination with a properly designed high-current output stage and adequate thermal management.
2) The LME49830 was mainly designed for output stages using MOSFET device, but may be used with other output stage device types as well such as BJT.
3) The LME49830 has a maximum ¡§DC bias voltage of 16 V¡¨ for output stage. In other word, it can be used for any MOSFET device, but providing that not to exceed this limit.
4) The LME49830 can be configured with a number of different output stage topologies such as output stage with 2 or 3 transistors paralleled per side.
5) The following MOSFET devices were tested and recommended by National Semiconductor:
=> Toshiba: 2SK1530/ 2SJ201
=> IR: IRFP240/ IRFP9240
=> Renesas: 2SK1058/ 2SJ162
Remark: I am confident that you read my English without problem because Chinese people in Hong Kong have no language barrier in English. In fact, I have lived in North America for some time before.
1) The LME49830 IC in theory can provide output power levels in excess 1 kW in combination with a properly designed high-current output stage and adequate thermal management.
2) The LME49830 was mainly designed for output stages using MOSFET device, but may be used with other output stage device types as well such as BJT.
3) The LME49830 has a maximum ¡§DC bias voltage of 16 V¡¨ for output stage. In other word, it can be used for any MOSFET device, but providing that not to exceed this limit.
4) The LME49830 can be configured with a number of different output stage topologies such as output stage with 2 or 3 transistors paralleled per side.
5) The following MOSFET devices were tested and recommended by National Semiconductor:
=> Toshiba: 2SK1530/ 2SJ201
=> IR: IRFP240/ IRFP9240
=> Renesas: 2SK1058/ 2SJ162
Remark: I am confident that you read my English without problem because Chinese people in Hong Kong have no language barrier in English. In fact, I have lived in North America for some time before.
Here are my understandings:
1) MOSFET or BJT, which is better for audio amplification?
=> My answer is MOSFET. It is a personal preference of musical sound produced by MOSFET. No argument if you like BJT.
2) Vertical (HEXFET) or Lateral MOSFET, which is better for audio amplification?
=> My answer is Lateral MOSFET. Lateral MOSFET suits for audio application while Vertical MOSFET (HEXFET) primarily aims at switching application (ON/OFF). But, all Lateral MOSFET, Vertical MOSFET and BJT are applicable for audio amplification application. Still works fine!
3) Classification of MOSFET applied and tested by National Semiconductor in LME49830?
=> Lateral MOSFET: Toshiba (2SK1530/2SJ201), Renesas (2SK1058/2SJ162)
=> Vertical MOSFET (HEXFET): International Rectifier (IRFP240/IRFP9240)
4) My target DIY Class AB power amplifier module:
=> Class AB Driver: LME49830 (mono design)
=> Output stage: Lateral MOSFET. e.g. Toshiba 2SK1530/2SJ201 (alternative if no choice: HEXFET)
=> Output stage topologies: 2 pairs or more MOSFET per channel (alternative if no choice: accept one pair per channel)
=> Output power: 200W/ 8ohm per channel (alternative if no choice: 100W/ 8ohm)
Feel free to comment if my understanding is incorrect!
Remark: I am still waiting for Monkey (CP Design) for development of captioned module at http://www.chip-pcb.cn
1) MOSFET or BJT, which is better for audio amplification?
=> My answer is MOSFET. It is a personal preference of musical sound produced by MOSFET. No argument if you like BJT.
2) Vertical (HEXFET) or Lateral MOSFET, which is better for audio amplification?
=> My answer is Lateral MOSFET. Lateral MOSFET suits for audio application while Vertical MOSFET (HEXFET) primarily aims at switching application (ON/OFF). But, all Lateral MOSFET, Vertical MOSFET and BJT are applicable for audio amplification application. Still works fine!
3) Classification of MOSFET applied and tested by National Semiconductor in LME49830?
=> Lateral MOSFET: Toshiba (2SK1530/2SJ201), Renesas (2SK1058/2SJ162)
=> Vertical MOSFET (HEXFET): International Rectifier (IRFP240/IRFP9240)
4) My target DIY Class AB power amplifier module:
=> Class AB Driver: LME49830 (mono design)
=> Output stage: Lateral MOSFET. e.g. Toshiba 2SK1530/2SJ201 (alternative if no choice: HEXFET)
=> Output stage topologies: 2 pairs or more MOSFET per channel (alternative if no choice: accept one pair per channel)
=> Output power: 200W/ 8ohm per channel (alternative if no choice: 100W/ 8ohm)
Feel free to comment if my understanding is incorrect!
Remark: I am still waiting for Monkey (CP Design) for development of captioned module at http://www.chip-pcb.cn
AndrewT said:
1) Toshiba 2SK1530/2SJ201 is well known to be very good for audio application. Noted that it is a vertial MOSFET, but not lateral. Where is your source to support it?
2) My requirement of power output should make sense and match with MOSFET output stage capability. For 100W, 1 or 2 pair(s) MOSFET is sufficient. For 200W, may be 3 or more MOSFET are required.
The Toshiba FETs are 150W rated (200V, 12A). Throwing the numbers into some calculations there should not be any problem for 200W into 8 ohm with the V/I phase 35 degrees. The power supply voltage should be around +/-65V and 2 devices per side (4 total in output stage). Now if the Renesas FETs are used under the same conditions the safe V/I phase is only 15 degrees as they are 100W rated devices. The IR FETs are similar in rating to the Toshiba so only the Renesas FETs would need to be increased in number, maybe. If this is wrong then I'd love to learn how to calculate safe operation so please share how you do it.
-SL
-SL
2SK1530/2SJ201
I did not say that they were Lateral MOSFET.
Re-read my post
But, in a way, if semantics is so important
I will start keeping a keen eye of
how you express every little detail in our forum.
Word for word .. marked well by Lineup.
And I wil post comments to undermine your diy audio authority, too
Now if National by some classification call 2SK1530/2SJ201 as being 'lateral type',
I prefer to rely on them first hand.
At least THEY should be Engineers with Papers of taken degrees.
Otherwise I would fire them right away
.. if I was Managing Director of National Semiconductor little company.
Secondly I can consider the private knowledge & diy amateurs opinions of people at this forum.
I have no idea what education or background all have here.
My own background I know
Lineup
Bensen's excel spreadsheet shows how to plot the SOAR with the load lines for FETs.
http://www.diyaudio.com/forums/showthread.php?postid=644221#post644221
My modified version also includes temperature derating and corrects a couple of minor errors.
Toshiba datasheet shows 2sj200 as 120W device increasing to 180W for a 100mS single non repetitive pulse @ Tc=25degC.
Most designers will allow for at least 45degree phase angle for the load and I usually adopt a severe phase angle of 60degrees to ensure reliability.
Look at capacitance, admittance and positive temp coef below 9A+.
They are all indicators that this does not mimic Lateral characteristics.
The one parameter that does look similar is Vgs, and maybe this is what has been tweaked that makes it more suitable for audio.
Vgs of Laterals & j200/201 and Vbe of BJTs are all of a similar level.
This makes me wonder if this lowish Vbe/Vgs is the important factor in designing a semiconductor for audio use.
http://www.diyaudio.com/forums/showthread.php?postid=644221#post644221
My modified version also includes temperature derating and corrects a couple of minor errors.
Toshiba datasheet shows 2sj200 as 120W device increasing to 180W for a 100mS single non repetitive pulse @ Tc=25degC.
Most designers will allow for at least 45degree phase angle for the load and I usually adopt a severe phase angle of 60degrees to ensure reliability.
Look at capacitance, admittance and positive temp coef below 9A+.
They are all indicators that this does not mimic Lateral characteristics.
The one parameter that does look similar is Vgs, and maybe this is what has been tweaked that makes it more suitable for audio.
Vgs of Laterals & j200/201 and Vbe of BJTs are all of a similar level.
This makes me wonder if this lowish Vbe/Vgs is the important factor in designing a semiconductor for audio use.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.