I read the data sheets of these transistor. I cannot tell why people favor BC550 and BC560 over the KSxxxx for the IPS differential pair. I looked at the max voltage, beta and input/output capacitance. KSxxxx are slightly better. What else do I need to look for?
Thanks
Thanks
operation near saturation.
Some of the BC series are way out in performance that others cannot match.
Price and availability, at least here in europe.
For transistor <60Vce0, the BC series cover many of our needs without having to stock dozens of others.
I now have dozens of others, at 2* to 20* the cost of bc550c bc546b but still use the BC when they suit.
Some of the BC series are way out in performance that others cannot match.
Price and availability, at least here in europe.
For transistor <60Vce0, the BC series cover many of our needs without having to stock dozens of others.
I now have dozens of others, at 2* to 20* the cost of bc550c bc546b but still use the BC when they suit.
Thanks,
In US, they all about the same price......dirt cheap!!! What difference does it make in saturation? They too, look close.
I really want to know what I should look for in a transistor other than fT, VCEO, beta and all the capacitance. The ones I posted are so close.
One thing I do notice, the pin out is different and BC with E, B C are easier to layout. The KS are E, C, B that B cross with either E or C in traces. That is a big plus in my book.
In US, they all about the same price......dirt cheap!!! What difference does it make in saturation? They too, look close.
I really want to know what I should look for in a transistor other than fT, VCEO, beta and all the capacitance. The ones I posted are so close.
One thing I do notice, the pin out is different and BC with E, B C are easier to layout. The KS are E, C, B that B cross with either E or C in traces. That is a big plus in my book.
Last edited:
Unless they are specials, all 2sa/c have the same pin out order.
But watch To126, it is reversed from To220
But watch To126, it is reversed from To220
Yeh, I don't like it. It does make a difference in layout as for 2 layer, crossing pin is a big deal. I am trying to have a ground plane for the board except at points that require low capacitance.
The issue with the lower Vceo and lower relative Gm devices might be a lower Early voltage figure. This figure is related to a specific 'kink' in the Gm transfer curve and is dependent on certain characteristics of the transistor. For the IPS, Vceo is generally kept relativity constant above or below this voltage so it is not an issue. For a VAS transistor that will be traversing this 'Early' region of operation, it may not be as suitable. Remember feedback does not 'fix' everything as is some rumors would have you believe.🙄
Early Voltage...
Early Voltage...
Last edited:
This argument actually favors KSC1845 and KSA992 as they are 120V devices vs 80V or under for the BCs.
So if I extend your argument, I should choose BJT that has higher VCEO as possible for power BJT?
Last edited:
I usually use MPSA42/92 for small signal use.
I use MJE340/350 for medium signal use and drivers.
For more powerful drivers I use mje15030 and mje15031.
I use MJE340/350 for medium signal use and drivers.
For more powerful drivers I use mje15030 and mje15031.
Bc550/560 - baxandall superpairs , input pairs that are already cascoded ,
active devices on a cascoded VAS , actually a better CM (mirror).
Ksa/c 992/1845 preferred low Ic cascode , current source , level shifter
maybe a input pair that is not cascoded or current feedback based..
Ksa/c 1381/3503 (mje340/350) - main VAS cascode/ main VAS CCS ,
predrivers for EF3. Don't use these anymore for the VAS actives , always
Cascoded.
Mje15033/32 good "industrial strength" driver ... up to 3 pair OP.
Sanken sa1294/sc3263 for bigger amps , other Sanken
to-3p's are <200pF Cob/ >50mhz ft (better than the MJE's).
Many other good choices , this is what the device papers are for.
They all ask "can I use X", and they could of compared the BOM
device with what they intend on using ??
OS
OS
active devices on a cascoded VAS , actually a better CM (mirror).
Ksa/c 992/1845 preferred low Ic cascode , current source , level shifter
maybe a input pair that is not cascoded or current feedback based..
Ksa/c 1381/3503 (mje340/350) - main VAS cascode/ main VAS CCS ,
predrivers for EF3. Don't use these anymore for the VAS actives , always
Cascoded.
Mje15033/32 good "industrial strength" driver ... up to 3 pair OP.
Sanken sa1294/sc3263 for bigger amps , other Sanken
to-3p's are <200pF Cob/ >50mhz ft (better than the MJE's).
Many other good choices , this is what the device papers are for.
They all ask "can I use X", and they could of compared the BOM
device with what they intend on using ??
OS
OS
Do you mean BC550/560 are better if I do cascode? Why?
What is the reason don't use these
BC550/650 perform much better at the low Vceo of a 12V cascoded
LTP. I have both types LTP , the BC's are quieter (and have higher Hfe).
If you cascode the typical "blameless" the current mirrors are also low
voltage - BCxxx performs better here , too.
On a cascoded VAS , also...lower Vceo for the active device , BCxxx rules
the day here , as well.
On "enhanced " VAS's (like a baxandall) "quasi saturation" of the BCxxx
is absolutely required for smooth operation.
Proper device for proper app.
edit - 2 most common VAS's - most stable and lowest THD , as well (below 1/2)
First one cancels the early effect , second one cancel's both early effect and Ccb ...
only use the second on a low Z CFA compensation setup.
OS
Attachments
Last edited:
Another alternative for the popular KSC1845 and KSA992 (TO92) and Ksa/c 1381/3503 (TO126) are the KSA1220A and KSA2690A both TO126.
The non "A" version is 120V and the "A" version 160V
They are available in selected gain ranges with the "Y" suffix being 160-320.
FT is 175MHz for the PNP with 26pF Cob and 155MHz and 19pF Cob.
While their Cob is higher than the Ksa/c 1381/3503 I have not found any problems with that.
I have used the MJE15030/31 for decades but the MJE15032/33 seems like a logical upgrade for the drivers.
Ron.
The non "A" version is 120V and the "A" version 160V
They are available in selected gain ranges with the "Y" suffix being 160-320.
FT is 175MHz for the PNP with 26pF Cob and 155MHz and 19pF Cob.
While their Cob is higher than the Ksa/c 1381/3503 I have not found any problems with that.
I have used the MJE15030/31 for decades but the MJE15032/33 seems like a logical upgrade for the drivers.
Ron.
Great, I am going to use BCs for all low voltage application as the bottom of the cascode.
I don't follow what is baxandall.
I looked at your schematic, basically what you meant is using BC550/560 for VAS, but use 2SAxxxx and 2SCxxxx to take the high voltage and the power?
Thanks
OK, Thanks guys, I think it's pretty much settle, BC550/560 for LV and 2SA/2SC for pre-driver and Vbe multiplier.
Now I am getting to the power BJT and the driver (2EF)
I am looking at between 2SA1943/2SC5200 vs MJW3281/MJW1302. MJW are 230V vs 2SC/2SA are 250V. BUT MJW is more expensive. I tend to go with 2SA and 2SC. What do you guys think?
For drivers in TO-220, I choose 2SC4883 and 2SA1930. Tell me what you guys think. They are 150V and 180V device, but this is for emitter follower only.
Thanks
Now I am getting to the power BJT and the driver (2EF)
I am looking at between 2SA1943/2SC5200 vs MJW3281/MJW1302. MJW are 230V vs 2SC/2SA are 250V. BUT MJW is more expensive. I tend to go with 2SA and 2SC. What do you guys think?
For drivers in TO-220, I choose 2SC4883 and 2SA1930. Tell me what you guys think. They are 150V and 180V device, but this is for emitter follower only.
Thanks
I'm actually listening to the 2'nd VAS at 200w peaks now.
nearly brings tears to my eyes .... NO OEM even comes close to either
"enhanced VAS" running through a EF3.
This is what the "audio phools" pay 4 K$ for.
Next level beyond the basic concept Cordell/self examples.
Edit - just build it and and you will say WTF/WOW !
PS - I'm even amazed by the improvement beyond what can be bought.
Edit 2 - "baxandall" is the second posted VAS with the dual BCxxx's on each
rail. (you have not seen that one in Bob's book)
OS
nearly brings tears to my eyes .... NO OEM even comes close to either
"enhanced VAS" running through a EF3.
This is what the "audio phools" pay 4 K$ for.
Next level beyond the basic concept Cordell/self examples.
Edit - just build it and and you will say WTF/WOW !
PS - I'm even amazed by the improvement beyond what can be bought.
Edit 2 - "baxandall" is the second posted VAS with the dual BCxxx's on each
rail. (you have not seen that one in Bob's book)
OS
Last edited:
Beyond the book , huh ! 😀
Post # 4729 on the http://www.diyaudio.com/forums/solid-state/248105-slewmaster-cfa-vs-vfa-rumble.html
slewmaster thread has what I listen to now.
Kypton C is the "alien" CFA input stage combined
with that "Baxandall" VAS. NOT in Bob's book!!
(only on my "channel")...
OS
Post # 4729 on the http://www.diyaudio.com/forums/solid-state/248105-slewmaster-cfa-vs-vfa-rumble.html
slewmaster thread has what I listen to now.
Kypton C is the "alien" CFA input stage combined
with that "Baxandall" VAS. NOT in Bob's book!!
(only on my "channel")...
OS
2sa1943/c5200 have very poor SOA compared to similar 150W devices.
They do NOT suit amplifiers with higher supply rail voltage.
I would limit them to <<+-60Vdc preferably <+-50Vdc
Compare the current ability of a 1943/5200 @ 50Vce & 70Vce to a 3281/1302 & 4281/4302 @ same Voltages.
For a robust 150W device, which is very much slower, Toshiba 2n3773, guaranteed by 100% testing for 1.5A @ 100Vce
For really high power look at mj15003/4 & mj21193/4
These three show what is possible and how much you give up when you move to the fast devices.
Last edited:
I only do +/-40V rail. I bought and received the 625VA 30V-30V toroidal transformer we talked about. So I am good.
Besides, I am using 5 pairs to get the quiescent current up at 200mA per pair to get more class A power. I don't think thermal is a problem for me.
- Status
- Not open for further replies.