High Voltage BjT's for Pre-driver stages of High Power Amplifier - Overview

Jump to Latest
Ian Finch

Ian Finch

Member
Joined 2010
Paid Member
MrCRC said:
....MJE243/MJE253.....MJE15034/MJE15035
Click to expand...
These are actually driver transistors for currents up to 4 and 8 amps Ic max. MJE243/253 are also only relatively low voltage types.

Predrivers, which you find in triple EF output stages, are usually much lower current and capacitance. e.g. KSA1381/C3503 as wrenchone suggests or even some TO92 types such as MPSA42/92.
 
de Ocampo

de Ocampo

Member
Joined 2010
Hi,

Search box brought me here while in quest for spice model for Toshiba's 2SC2383/SA1013 devices. Most probably obselete but still available in this part of the world.

Anyone out there keeps a spice model for these devices?

Anything similar in specs [in spice model] could also be a good replacement at least in simulation works, perhaps another Toshiba pair 2SC2235/SA965.

Thanks!
 
de Ocampo

de Ocampo

Member
Joined 2010
Found it!....after patiently following endless link I found the 2sc2383/2sa1013 in the models attached. It was an extract from OrCAD lib simulator and was posted at the Software tools section. I have tested it in LTSpice and it seemed to work. :)
 

Attachments

  • japanese_bipolar.lib.txt
    165.3 KB · Views: 161
keantoken

keantoken

Member
Joined 2006
Cje is wrong, you can see they used 5pF for every transistor. Also Vaf is 100. It doesn't look like anyone really knew how to make good models in 1998, maybe a handful of people. If you design an amp in simulation with these models, it will probably not work if you make it in real life.
 
Last edited:
de Ocampo

de Ocampo

Member
Joined 2010
...by the way this is the reason why I eagerly wanted to do simulation with those Toshiba devices.It was used in VAS.
 

Attachments

  • Koda_amplifier.png
    Koda_amplifier.png
    680.7 KB · Views: 622
B

Bob Cordell

Member
Joined 2006
cordellaudio.com
keantoken said:
Cje is wrong, you can see they used 5pF for every transistor. Also Vaf is 100. It doesn't look like anyone really knew how to make good models in 1998, maybe a handful of people. If you design an amp in simulation with these models, it will probably not work if you make it in real life.
Click to expand...

Typical for manufacturer's models.

Cheers,
Bob
 
de Ocampo

de Ocampo

Member
Joined 2010
keantoken said:
Cje is wrong, you can see they used 5pF for every transistor. Also Vaf is 100. It doesn't look like anyone really knew how to make good models in 1998, maybe a handful of people. If you design an amp in simulation with these models, it will probably not work if you make it in real life.
Click to expand...


Maestro keantoken,

I was confused by these entries in a copy pasted spice models for TIP41C/42C...[while scanning the data entries in Notepad++]. Is the TIP42C IKR a valid data?...or perhaps the copy pasted data contains error. :(

TIP41C IKR=3.54059
TIP42C IKR=10 4

Sa muli,
Albert
 
Last edited:
keantoken

keantoken

Member
Joined 2006
The 4 at the end is wrong, it must be a copying error. However IKR only matters when the transistor is biased backwards, IE collector and emitter swapped. So it probably won't change anything. IKR=10 looks like a generic value, which doesn't speak well to the quality of the rest of the model.
 
tiefbassuebertr

tiefbassuebertr

Member
Joined 2005
tiefbassuebertragung.de
Replacement for Toshiba's 2SA1932/2SC5174 wanted (230V/1W-outline similar to ISO220)

For this transistors I want a replacement:
http://www.kynix.com/uploadfiles/pdf/2SC5174.pdf
2SA1932 pdf, 2SA1932 description, 2SA1932 datasheets, 2SA1932 view ::: ALLDATASHEET :::

The application therefore is the first transistor of an output triplet in a power amp (power follower stage), which is in use on a RB1090 from Rotel - go to
ROTEL RB 1090 Service Manual free download, schematics, eeprom, repair info for electronics
The emitter resistor is 1000R - thus the necessary current delivery capability must be actually low.
Rotel had nevertheless decided for a more large transistor outline (above TO126/TO220). Why?
 
B

Bob Cordell

Member
Joined 2006
cordellaudio.com
tiefbassuebertr said:
For this transistors I want a replacement:
http://www.kynix.com/uploadfiles/pdf/2SC5174.pdf
2SA1932 pdf, 2SA1932 description, 2SA1932 datasheets, 2SA1932 view ::: ALLDATASHEET :::

The application therefore is the first transistor of an output triplet in a power amp (power follower stage), which is in use on a RB1090 from Rotel - go to
ROTEL RB 1090 Service Manual free download, schematics, eeprom, repair info for electronics
The emitter resistor is 1000R - thus the necessary current delivery capability must be actually low.
Rotel had nevertheless decided for a more large transistor outline (above TO126/TO220). Why?
Click to expand...

I like to use TO126 devices for the first stage of a Triple output stage. Do not under-estimate the current delivery requirements and safe area requirements of the pre-driver under high-output-current conditions, especially considering beta degradation in the driver and output transistors. Also, do not overlook the stressful conditions of a short-circuit condition during the brief time before short-circuit protection comes into play.

I like to run the pre-drivers at 20mA in a larger amplifier (say, >150W). If the no-load rails are 65V, the dissipation is 1.3W. Also, the TO126 package makes mounting to a heat sink easy.

Also bear in mind thermal bias stability in the Triple, where 3 Vbe are in series, so junction temperature changes, especially during heat-up, can significantly affect the idle bias. For this reason, I like to keep the junctions from heating up to much; this in turn leads to the desire to heat-sink the devices.

I also like to use TO-126 for the VAS.

Cheers,
Bob
 
tiefbassuebertr

tiefbassuebertr

Member
Joined 2005
tiefbassuebertragung.de
Hi Bob, thank you for your advices. And I have found a better available replacement part - obviously from the same process: the 2SA1837/2SC4793 in an ordinary TO220-ISO outline:
http://www.adpslab.com/images/AdcomGFA-555MKII/datasheets/2sc4793.pdf
http://www.mouser.com/ds/2/408/toshiba_2SA1837-587291.pdf

Also the FZT957/FZT857 - go to
https://www.diodes.com/assets/Datasheets/FZT957.pdf
could be a replacement, if I use a little copper heatsink.

Toshiba's special outline "2-10T1A" from the used 2SC5174/2SA1932 under
http://www.kynix.com/uploadfiles/pdf/2SC5174.pdf
isn't necessary, so I think.

What happens, if the transit frequency is less than 100MHz ?
- go to the MJE15034/15035
https://www.onsemi.com/pub/Collateral/MJE15034-D.PDF
 
B

Bob Cordell

Member
Joined 2006
cordellaudio.com
tiefbassuebertr said:
Hi Bob, thank you for your advices. And I have found a better available replacement part - obviously from the same process: the 2SA1837/2SC4793 in an ordinary TO220-ISO outline:
http://www.adpslab.com/images/AdcomGFA-555MKII/datasheets/2sc4793.pdf
http://www.mouser.com/ds/2/408/toshiba_2SA1837-587291.pdf

Also the FZT957/FZT857 - go to
https://www.diodes.com/assets/Datasheets/FZT957.pdf
could be a replacement, if I use a little copper heatsink.

Toshiba's special outline "2-10T1A" from the used 2SC5174/2SA1932 under
http://www.kynix.com/uploadfiles/pdf/2SC5174.pdf
isn't necessary, so I think.

What happens, if the transit frequency is less than 100MHz ?
- go to the MJE15034/15035
https://www.onsemi.com/pub/Collateral/MJE15034-D.PDF
Click to expand...

30 MHz for a driver is OK, in my opinion. Just remember to run it with a healthy bias current, perhaps on the order of 50mA or more.

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