Complementary med pwr transistors?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
jan.didden

jan.didden

AX tech editor
Joined 2002
linearaudio.net
Paid Member
Hi all,

I am looking for complementary med power transistors, min specs 100V min Vce, Icont min 100mA, Pav min 3W, prefarably no SMD, TO220 would be fine, I've got enough heatsink real estate.

Since I grew up with BD139 (which doesn't meet the Vce spec), and the BD239 (which has rather low Hfe), I'm a bit out of touch with the newest devices. They will be used in current mirrors in my new power amp. Any recommnedations?

Jan Didden
 
Tube_Dude

Tube_Dude

Member
Joined 2002
janneman said:


Since I grew up with BD139 (which doesn't meet the Vce spec), and the BD239 (which has rather low Hfe), I'm a bit out of touch with the newest devices. They will be used in current mirrors in my new power amp. Any recommnedations?
Click to expand...

Hi

Hitachi 2SB649 / 2SD669
Case TO-126 - 180 Volts- collector current 1,5A-Power dissipation 20 W- hFE (60-320)- fT 140 Mhz

or Toshiba 2SC3298B / 2SA1306B
200Volts- 1,5A- 20W- hFE (70-240)- fT 100 Mhz

or Toshiba 2SA1837 / 2SC4793
230 V - 1 A- 20W- hFE (100- 320)- fT 70 Mhz

Regards
 
sam9

sam9

Member
Joined 2002
drachen-audio.com
TIP31c 7 TIP32c. I think some other "TIP" series excluding the Darlington's would fit the requirements. A bit old fashioned but they do the job.

One advantage of the TO-220 package vs. TO-126 is for some applications the larger collector plate on the back provides sufficient dheat dissapation without adding a clip on sink.
 
andy_c

andy_c

Banned
Joined 2003
andycpublic.50webs.com/
janneman said:
Yes, looks good. I will probably need to buy a 100 or so and select them, because they are for current mirrors, and they need to swing close to the supply so I don't want to use emitter degeneration to force an Ic match. I have no idea how close they would be in Vbe in a same-batch.
Click to expand...

That's a tough requirement! How different will the VCE values be between the two mirror transistors? If they are very different between the two, you'll get tracking errors due to the Early effect, as well as tracking errors due to the junction temperature mismatch between the two (since IS is a strong function of temperature).
 
sam9

sam9

Member
Joined 2002
drachen-audio.com
" Ft and *very* low Hfe above a 100 mA Ic. "

True enough. The fT is about the same as the older MJE15xxx power devices so they may work together well. I heard from a service tech that higher fT is not always beneficial. Replacing drivers and output devices in some older amps with higher fT devices sometimes just brings on scillations that don't ocurr with the orginal "old fashioned" components.

I take to heart the H&H dictim that Hfe is a "bad" number due the wide min/max spec. Also that any design that requires anything above the minimum on the datasheet is questionable. Personally I have found certain "high beta" transistors exibit such a wide variance piece-to-piece that they frequently measure lower than some supposedly inferior components, yet all is still within the datasheet spec.
 
M

mandat

Member
Joined 2003
Try Zetex - quality is unquestionable but prices...
http://www.zetex.com/3.0/3-3-2.asp?scatid=13

BD 139 N 80V 1,5A 12,5W 50MHz analogues:

2SC2078 80V 3A 10W 150 MHz
2SC1431 110V 2A 23W 80 MHz
2SC2238 160V 1,5A 25W 100 MHz
2SC2275 120V 1.5A 25W 200 MHz
2SC3421 120V 1A 10W 120 MHz
2SC3944 150V 1A 40W 300 MHz

BD140 P 80V 1.5A 12.5W 50 MHz analogues:

2SA1225 160V 1,5A 15W 100 MHz
2SA1220A120V 1.2A 20W 160 MHz
2SA1011 160v 1.5A 25W 120 MHz
2SA1111 150V 1A 20W 200 MHz
2SA1358 120V 1A 10W 120 MHz
2SA1535 180V 1A 40W 200 MHz

http://www.necel.com/discrete/images/pdf_e/TC-3586.pdf
http://www.necel.com/discrete/images/pdf_e/TC-3764.pdf
http://www.necel.com/discrete/images/pdf_e/TC-3763.pdf

There are 2SB768/2SD1033 150V 2A in TO251 case 2SB546A/2SD401 in TO220 case
 
jan.didden

jan.didden

AX tech editor
Joined 2002
linearaudio.net
Paid Member
andy_c said:


That's a tough requirement! How different will the VCE values be between the two mirror transistors? If they are very different between the two, you'll get tracking errors due to the Early effect, as well as tracking errors due to the junction temperature mismatch between the two (since IS is a strong function of temperature).
Click to expand...


True, hadn't thought about that. Damn! I need to set a current within say 5%, AND I want to be able to swing close to the supply. Maybe I should forget about a current mirror and try a feedback pair, you know, put a resistor in the emitter, sense that voltage and control the base voltage. But that would mean I lose at least .6V swing. Maybe I got to accept that.


Jan Didden
 
andy_c

andy_c

Banned
Joined 2003
andycpublic.50webs.com/
janneman said:
True, hadn't thought about that. Damn! I need to set a current within say 5%, AND I want to be able to swing close to the supply. Maybe I should forget about a current mirror and try a feedback pair, you know, put a resistor in the emitter, sense that voltage and control the base voltage. But that would mean I lose at least .6V swing. Maybe I got to accept that.
Click to expand...

If you go with a 3-transistor Wilson current mirror, only the common-base device needs to have a high voltage rating, and the two mirror devices can be low-voltage devices. This eliminates the Early voltage current tracking error as well, as the low-voltage devices have almost equal VCB. If you only need NPN devices, you could use a monolithic transistor array for the low-voltage devices to improve the matching.
 
jan.didden

jan.didden

AX tech editor
Joined 2002
linearaudio.net
Paid Member
Pooge,

Thanks, I think fig b is what Andy alluded to. I found a similar circuit in Walt Jung's IC Array Cookbook:




Jan Didden
 

Attachments

  • isource.jpg
    isource.jpg
    11.8 KB · Views: 171
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom