UrSv,
You are quite right the datasheet says "High power amplifier".
I had already forgotten that and only looked at the actual data
when rechecking the datasheet. Sorry about that. Since no
fT value was specified, but all info about rise and fall times etc.
and also a switching test circuit was supplied, I inferred that
it was a switching device. The RDSon value is missing in the
datasheet, just as you observerd. Incidentally, it is specified in
the general power transistor catalogue
http://www.semicon.toshiba.co.jp/eng/prd/comminf/pdf/217_242.pdf (p. 237 or p. 21 within this particular document).
which simply lists all MOSFETs as general purpose. I am not
an expert on the subject, but since MOSFETs don't have any
problems with minority carriers, perhaps there is no real
distinction between linear and switching devices, so the usage
and labelling is more arbitrary than for BJTs? I think I have
seen quite a number of amplifiers actually using switching MOSFETs, but maybe I am wrong there?
Regarding "PowerMOS" I simply meant that as a general term
for all power MOSFETs, but perhaps it refers to a specific type
or brand (it does sound a bit familiar).
BTW, please note that I don't mean to argue with you. I am just
curious to find out what these devices really are.
You are quite right the datasheet says "High power amplifier".
I had already forgotten that and only looked at the actual data
when rechecking the datasheet. Sorry about that. Since no
fT value was specified, but all info about rise and fall times etc.
and also a switching test circuit was supplied, I inferred that
it was a switching device. The RDSon value is missing in the
datasheet, just as you observerd. Incidentally, it is specified in
the general power transistor catalogue
http://www.semicon.toshiba.co.jp/eng/prd/comminf/pdf/217_242.pdf (p. 237 or p. 21 within this particular document).
which simply lists all MOSFETs as general purpose. I am not
an expert on the subject, but since MOSFETs don't have any
problems with minority carriers, perhaps there is no real
distinction between linear and switching devices, so the usage
and labelling is more arbitrary than for BJTs? I think I have
seen quite a number of amplifiers actually using switching MOSFETs, but maybe I am wrong there?
Regarding "PowerMOS" I simply meant that as a general term
for all power MOSFETs, but perhaps it refers to a specific type
or brand (it does sound a bit familiar).
BTW, please note that I don't mean to argue with you. I am just
curious to find out what these devices really are.
Christer,
Sorry of the respons came across as a start of an argument. It was actually nothing more than a shortlist of why I believe it is for Audio and possibly also a glimpse at the lack of "definitive" facts on the matter.
Trivia 1: Our Swedish homeboy M Bladelius uses them in the Advantage product line.
I am definitively not the man to say if a MOS switching transistor is better or worse than a transistor specifically designed for Audio but as you say switching type transistors seem to be used quite often. However I think the types are generally the more general types rather than the most specialised switching types.
/UrSv
Sorry of the respons came across as a start of an argument. It was actually nothing more than a shortlist of why I believe it is for Audio and possibly also a glimpse at the lack of "definitive" facts on the matter.
Trivia 1: Our Swedish homeboy M Bladelius uses them in the Advantage product line.
I am definitively not the man to say if a MOS switching transistor is better or worse than a transistor specifically designed for Audio but as you say switching type transistors seem to be used quite often. However I think the types are generally the more general types rather than the most specialised switching types.
/UrSv
For What its worth, I have used the Toshiba 1529/200 pair in a design. I do like these FETS. I am planning on using the Hitachi 1058/162s next.
So for clarification, Evolve Amplifiers site in Japan list these (1529, 1530, 200, 201) as designed for Audio.
Also, they ARE lateral, but the neg. temperature coef. doesn't kick in till 6 Amps for the 1529/200 and about 8 Amps for the 1530/201. This defeats the positive aspects of the negative temp. coef. I.E. temp compensation is needed (via Vbe multipier).
One other thing of note is that the voltage bias to turn them on (for class B) is lower than for the IRF type devices. The toshibas needed a bias of about 1.4 - 1.6 Volts each to turn them on, and I found the IRF 540/9540 to need > 2 Volts. I suspect that these devices will run a bit cooler with regards to this, but I havn't actually tested to see.
I also think that Holten's 'Scalable Power Amp' could be modified to use any device keeping in mind the Biasing and temp compensation circuitry needed for the devices that you would use.
-Dozuki
So for clarification, Evolve Amplifiers site in Japan list these (1529, 1530, 200, 201) as designed for Audio.
Also, they ARE lateral, but the neg. temperature coef. doesn't kick in till 6 Amps for the 1529/200 and about 8 Amps for the 1530/201. This defeats the positive aspects of the negative temp. coef. I.E. temp compensation is needed (via Vbe multipier).
One other thing of note is that the voltage bias to turn them on (for class B) is lower than for the IRF type devices. The toshibas needed a bias of about 1.4 - 1.6 Volts each to turn them on, and I found the IRF 540/9540 to need > 2 Volts. I suspect that these devices will run a bit cooler with regards to this, but I havn't actually tested to see.
I also think that Holten's 'Scalable Power Amp' could be modified to use any device keeping in mind the Biasing and temp compensation circuitry needed for the devices that you would use.
-Dozuki
Dozuki is correct. Toshiba originally had an audio sub-brand called "Aurex", and at the beginning of the 1980s, Toshiba's semiconductor division introduced a pair of devices called 2SK405 and 2SJ115, and these were targeted for Aurex's audio power amps. The 2SK1529/1530 2SJ200/201 are second-generation audio power devices within the same pi-MOS lineup.
"Pi" refers to the current-flow pattern within the devices. These particular devices had many interesting characterstics about them, but I won't go into detail right now, as they are Out Of Production.
The 2SK1529/1530 2SJ200/201 need some thermal compensation measures, but not as comprehensive as would be required for a high-current switching MOSFET. A simple VBe multiplier, a thermal-coupled diode, or a thermistor should suffice. Of course, the amplifier designer is free to implement more elaborate schemes if he so chooses.
The Hitachi 2SK1056~1058 2SJ160~162 are electrically the spitting image of the older 2SK133~135 2SJ48~50 family, and can by used with a very simple resistor bias network.
regards, jonathan carr
"Pi" refers to the current-flow pattern within the devices. These particular devices had many interesting characterstics about them, but I won't go into detail right now, as they are Out Of Production.
The 2SK1529/1530 2SJ200/201 need some thermal compensation measures, but not as comprehensive as would be required for a high-current switching MOSFET. A simple VBe multiplier, a thermal-coupled diode, or a thermistor should suffice. Of course, the amplifier designer is free to implement more elaborate schemes if he so chooses.
The Hitachi 2SK1056~1058 2SJ160~162 are electrically the spitting image of the older 2SK133~135 2SJ48~50 family, and can by used with a very simple resistor bias network.
regards, jonathan carr
Christer said:
Note that fT is never listed in datasheets of FET's.
The interstings things are switch times, capacitances and the inductances (bond wires).
My idea about the advantage of lateral fet's is the almost constant gm, good for distortion. The disadvantage is the high output impedance.
1 A, 1 Ohms, lateral
1 A, 0.05 ohms, BJT
5 A, 1 Ohms, lateral
5 A, 0.005 ohms, BJT
What do you suppose makes them so linear?
Extremely short channel length devices have linear behavior, but I doubt that is how these things achieve it.
The reason Ft is never listed is that it is easily calculated from the input and output currents of a common source amplifier:
Iin = 2*pi*f*Cgs*Vin.
Iout = Gm*Vin.
in which Vin is the (small signal) amplitude of the input sinewave.
Ft is defined as unity current gain, ie Iin = Iout.
So Ft = Gm/(2*pi*Cgs)
For most MOSFETs this is too bias dependent, becasue Gm varies nearly linearly with Id. But these lateral ones are odd in that after about 1A of drain current, they become linear. Maybe it is velocity saturation as in short channel FETs after all.
Extremely short channel length devices have linear behavior, but I doubt that is how these things achieve it.
The reason Ft is never listed is that it is easily calculated from the input and output currents of a common source amplifier:
Iin = 2*pi*f*Cgs*Vin.
Iout = Gm*Vin.
in which Vin is the (small signal) amplitude of the input sinewave.
Ft is defined as unity current gain, ie Iin = Iout.
So Ft = Gm/(2*pi*Cgs)
For most MOSFETs this is too bias dependent, becasue Gm varies nearly linearly with Id. But these lateral ones are odd in that after about 1A of drain current, they become linear. Maybe it is velocity saturation as in short channel FETs after all.
Lateral Mosfet toshiba 2sk405 /2sJ115 substitude
As was stated in this forum the toshiba 2sk 1529 ,1530 - 2sj 200, 201 are the new versions for those older mosfets used in Aurex amps and Marantz , Luxman ect. with same pi and ......
But as not beeing an expert I looked at the website from toshiba and found more differences witch made me wonder now whats a better substitude. here is the list.
TO-3P(N) Package
TO-3P(N)
2SK1529 Nch 180 10 120 0.83 10 -
2SK3497 180 10 130 0.15 7.0 36
2SJ200 Pch -180 -10 120 0.83 -10 -
2SJ618 -180 -10 130 0.37 -7.0 35
TO-3P(L) Package
TO-3P(L)
2SK1530 Nch 200 12 150 0.625 10 -
2SJ201 Pch -200 -12 150 0.625 -10 -
as you can see there is talk of 3p
package and 3p(L) if that L stands for Lateral tat would implicate that only the 1530/201 are to be used.
On the other hand If I remember reading somewhere that the old version k405/J115 was 180 volt and 10A the K1529/J200 are a better match.
So how important is the L and may it also be 200v 12A instead.
When do these newer fets switch in if the older version would doe this at 1.5 volt what if thothe newer version does it only at 2v will the amp heat up more?
And the Hitachi-Renesas 2SJ162/2SK1058
Item Symbol Min Typ Max Unit Test conditions
As was stated in this forum the toshiba 2sk 1529 ,1530 - 2sj 200, 201 are the new versions for those older mosfets used in Aurex amps and Marantz , Luxman ect. with same pi and ......
But as not beeing an expert I looked at the website from toshiba and found more differences witch made me wonder now whats a better substitude. here is the list.
TO-3P(N) Package
TO-3P(N)
2SK1529 Nch 180 10 120 0.83 10 -
2SK3497 180 10 130 0.15 7.0 36
2SJ200 Pch -180 -10 120 0.83 -10 -
2SJ618 -180 -10 130 0.37 -7.0 35
TO-3P(L) Package
TO-3P(L)
2SK1530 Nch 200 12 150 0.625 10 -
2SJ201 Pch -200 -12 150 0.625 -10 -
as you can see there is talk of 3p
package and 3p(L) if that L stands for Lateral tat would implicate that only the 1530/201 are to be used.On the other hand If I remember reading somewhere that the old version k405/J115 was 180 volt and 10A the K1529/J200 are a better match.
So how important is the L and may it also be 200v 12A instead.
When do these newer fets switch in if the older version would doe this at 1.5 volt what if thothe newer version does it only at 2v will the amp heat up more?
And the Hitachi-Renesas 2SJ162/2SK1058
Item Symbol Min Typ Max Unit Test conditions
mosfets substitude K405/J115 carry on
somehow still writing and pushing a wrong button the thing went up on to forum while I was not finished. sorry.
next question is would it also be possible to change those
Toshiba K405/J115 for the
Hitachi - Renesas
2SK1058 /SJ 162 being 160V 7A
Gate to source breakdown voltage V(BR)GSS ±15 — — V IG = ±100 μA, VDS = 0
Gate to source cutoff voltage VGS(off) 0.15 — 1.45 V ID = 100 mA, VDS = 10 V
Drain to source saturation voltage VDS(sat) — — 12 V ID = 7 A, VGD = 0 *2
Forward transfer admittance |yfs| 0.7 1.0 1.4 S ID = 3 A, VDS = 10 V *2
Input capacitance Ciss — 600 — pF
Output capacitance Coss — 350 — pF
Reverse transfer capacitance Crss — 10 — pF
VGS = –5 V, VDS = 10 V,
f = 1 MHz
Turn-on time ton — 180 — ns
Turn-off time toff — 60 — ns
VDD = 20 V, ID = 4 A
Note: 2. Pulse test
ore
Magna Tec BUZ 900P/901P 160v/200v 8A ( not sure if lateral )
input capacitance 500
output capacitance 300
reverse transfer 10
turn on time 100
turn off time 50
many people are guessing about this and very few know how important the little tech. stuff mite be putting different than the original in your amp.
So perhaps it would be possible to give some explanation about how important are those things as being a lateral Fet and what if it has 200v 12A instead of the original 180v 10A
That this will produce more watts ok, but will it work what problems can be expected from the differences and what must not be to far of the original specs.
Thanks very much
somehow still writing and pushing a wrong button the thing went up on to forum while I was not finished. sorry.
next question is would it also be possible to change those
Toshiba K405/J115 for the
Hitachi - Renesas
2SK1058 /SJ 162 being 160V 7A
Gate to source breakdown voltage V(BR)GSS ±15 — — V IG = ±100 μA, VDS = 0
Gate to source cutoff voltage VGS(off) 0.15 — 1.45 V ID = 100 mA, VDS = 10 V
Drain to source saturation voltage VDS(sat) — — 12 V ID = 7 A, VGD = 0 *2
Forward transfer admittance |yfs| 0.7 1.0 1.4 S ID = 3 A, VDS = 10 V *2
Input capacitance Ciss — 600 — pF
Output capacitance Coss — 350 — pF
Reverse transfer capacitance Crss — 10 — pF
VGS = –5 V, VDS = 10 V,
f = 1 MHz
Turn-on time ton — 180 — ns
Turn-off time toff — 60 — ns
VDD = 20 V, ID = 4 A
Note: 2. Pulse test
ore
Magna Tec BUZ 900P/901P 160v/200v 8A ( not sure if lateral )
input capacitance 500
output capacitance 300
reverse transfer 10
turn on time 100
turn off time 50
many people are guessing about this and very few know how important the little tech. stuff mite be putting different than the original in your amp.
So perhaps it would be possible to give some explanation about how important are those things as being a lateral Fet and what if it has 200v 12A instead of the original 180v 10A
That this will produce more watts ok, but will it work what problems can be expected from the differences and what must not be to far of the original specs.
Thanks very much
Re: Lateral Mosfet toshiba 2sk405 /2sJ115 substitude
TO3P(L) is the package also known as TO264 while TO3P is the japanese style TO3P similar in size to the TO247. Nothing to do with what type of chip is inside. 2SC5200, 2SC3281, 2SA1943, 2SA1302 is also TO3P(L) package for example.
The Toshiba devices are not lateral. They are some kind of vertical type that is specially made for audio. I'm pretty sure Toshiba has never made any lateral power MOSFET:s.
It's easy to determine if a FET is lateral or not. Source is the case/middle pin on a lateral FET while drain is middle pin/case on a vertical FET. It can also be seen from the Id/Vgs curves.
Lateral FETs and vertical types are not directly interchangable. The different pinout needs to be accomodated and also the biasing needed is different.
The Magnatec BUZ types are lateral fets. (look at pinout and infliction point in Id/Vgs curve)
TO3P(L) is the package also known as TO264 while TO3P
is the japanese style TO3P similar in size to the TO247. Nothing to do with what type of chip is inside. 2SC5200, 2SC3281, 2SA1943, 2SA1302 is also TO3P(L) package for example.The Toshiba devices are not lateral. They are some kind of vertical type that is specially made for audio. I'm pretty sure Toshiba has never made any lateral power MOSFET:s.
It's easy to determine if a FET is lateral or not. Source is the case/middle pin on a lateral FET while drain is middle pin/case on a vertical FET. It can also be seen from the Id/Vgs curves.
Lateral FETs and vertical types are not directly interchangable. The different pinout needs to be accomodated and also the biasing needed is different.
The Magnatec BUZ types are lateral fets. (look at pinout and infliction point in Id/Vgs curve)
Re: mosfets substitude K405/J115 carry on
The Vds of 200V cf 180V is the maximum voltage rating. Using the higher rated device gives a higher margin of reliability but not necessarily any increase in performance if the 180V device is already adequate.
The Id of 12A cf 10A again is the device rating for maximum current. It has little bearing on the performance of the amplifier since the device never approaches max Id cold when operating into the specified load/s.
Again it will not give any extra power.
The higher Id rating may slightly increase resistance to damage in event of a short circuit of the output to ground, but, we don't need that do we?
no. the amp design as a whole determines the power output.Draber jorg said:
The Vds of 200V cf 180V is the maximum voltage rating. Using the higher rated device gives a higher margin of reliability but not necessarily any increase in performance if the 180V device is already adequate.
The Id of 12A cf 10A again is the device rating for maximum current. It has little bearing on the performance of the amplifier since the device never approaches max Id cold when operating into the specified load/s.
Again it will not give any extra power.
The higher Id rating may slightly increase resistance to damage in event of a short circuit of the output to ground, but, we don't need that do we?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.