Thank you Celestial. that makes sence.
Unfortunatelly, it looks like ksc1845 is not a 2sc2240 replacement noise-wise.
It's good for high source impedance and low current.
I'm looking for something with low noise for100-1000Ohm source and 2-3mA Ic.
2sc2545-2547 is OK, but also hard to find and quit slow with 2mA (less than 100MHz ft)
Unfortunatelly, it looks like ksc1845 is not a 2sc2240 replacement noise-wise.
It's good for high source impedance and low current.
I'm looking for something with low noise for100-1000Ohm source and 2-3mA Ic.
2sc2545-2547 is OK, but also hard to find and quit slow with 2mA (less than 100MHz ft)
The 2SC2240 had an Rbb around 30 ohm (from memory) you could look at using the ON Semi BC338-40's Rbb is around 30 ohm also (you'll need two BC338's in parallel to get near the performance of the 2SC2545 in terms of Rbb), voltage and current though are utterly different.
Avoid the Fairchild flavour (IMO) as it is not as good performance wise as the ON Semi part - compare the datasheets for yourself to see what I mean. In my experience the Fairchild part was noisier, slower(fT) and suffered greater Cob, Hfe was not as linear (I used to have access to an HP4156A Precision Semiconductor Parameter Analyzer with noise test setup - so can vouch for the performance of the BC338 parts from ON and Philips)
Although not intended as a low noise part, the older PHILIPS and ON SEMI BC337-40/BC327-40 parts worked well for me as an MC head preamp, using 8-10 devices in parallel. Compared to the 2SC3329, say, you've got a few trade-offs to consider.
Last edited:
Yes, i knew about 30 Ohms of 2240. It's low enough for me.
You seem to be expirienced with BJTs,
Do you have any information about the sanyo video BJTs? i'm talking about 2sa1370/1380 2sc3467/3502 ?
they look good for cascodes/common base VAS-out in power amp aplication, low Cob and good linearity. What about their noise? Can they be used in the front end too?
You seem to be expirienced with BJTs,
Do you have any information about the sanyo video BJTs? i'm talking about 2sa1370/1380 2sc3467/3502 ?
they look good for cascodes/common base VAS-out in power amp aplication, low Cob and good linearity. What about their noise? Can they be used in the front end too?
You might consider looking at the SANYO 2SA1016, 2SA1016K/ 2SC2362, 2SC2362K devices...They are in the same ball park as the 2SC2240 / 2SA970 devices, I'm not sure where the Rbb lies, I never actually put any of these on the HP 4156 to extract the Gummel-Poon parameters.
Attached is a datasheet for these parts.
I understand that ON Semi now owns the SANYO Semiconductor business so they might be easier to obtain outside Japan now - even in Japan, SANYO Semiconductor at the time I was there, was preoccupied with serving its own internal business units, external sales seemed to be a nice to have rather than a core function...
If you want a direct replacement for the TOSHIBA 2SC2240 then have a look at KEC's KTC3200 part - it has been recently discontinued but there may still be stocks around. I've used this in a product where a direct replacement was needed after TOSHIBA pulled out on an order...
So the people I was working for at the time were frantically running around trying to find surplus as production had ceased in the factory. Any how the KEC people turned up with the KTC3200 and the KTA1268 and they dropped right in the circuit quite nicely.
For some reason KEC went and pulled production of these parts last year - God knows why as with TOSHIBA and RENESAS withdrawing their small signal through hole parts you would have thought that KEC would have been looking to capitalise on this.
I didn't mention these parts earlier as they have been discontinued for 2 years now....
Try the SANYO parts and let us know how you go....We did contemplate using them, but their price wasn't as sharp at TOSHIBAs 2SC2240/2SA970 and I recall something about lead times and volumes being an issue?
The TOSHIBA parts used to be available from anyone of about 3 semiconductor sales houses we used to deal with - so the just in time production worked well with the TOSHIBA parts.
Last edited:
I probably need to clarify the KTC3200 - someone just mailed me saying this was a bad, bad, bad device - yes the early production runs didn't help KEC's reputation one bit.
There were issues with the early production of the KTC3200's, especially when used in a driver application with current approaching their SOA ratings - they tended to fail!
KEC at the time claimed that it was due to impurities - we'll never know: KEC then brought out the KTC3201 as a revised die/process - this worked fine.
However, I've never used the KTC3200 in driver applications, I have used it in small signal applications for diff pair and VAS stages and found out to be very reliable.
At the time when the KEC rep showed up with the KEC3200 at our office, it took us an afternoon and a few phone calls to discover that it was fine in the gain and diff pair stages, but keep it well clear of a driver stage where it had to push or pull current.
The KTC3201 we were told was the "fixed up" version!
There were issues with the early production of the KTC3200's, especially when used in a driver application with current approaching their SOA ratings - they tended to fail!
KEC at the time claimed that it was due to impurities - we'll never know: KEC then brought out the KTC3201 as a revised die/process - this worked fine.
However, I've never used the KTC3200 in driver applications, I have used it in small signal applications for diff pair and VAS stages and found out to be very reliable.
At the time when the KEC rep showed up with the KEC3200 at our office, it took us an afternoon and a few phone calls to discover that it was fine in the gain and diff pair stages, but keep it well clear of a driver stage where it had to push or pull current.
The KTC3201 we were told was the "fixed up" version!
Last edited:
From a quick inspection of the datasheets for the 2SC3467 and the 2SC3502, these parts don't look to be anywhere near the 2SC2240 in characteristic behavior.
At lower levels of Ic their GB product drops quite dramatically, their low voltage (Vce) performance isn't great, Cob is quite low though - these parts are really focused on one thing amplifying high frequency video signals to drive the colour guns in a CRT. They don't look suitable for a VAS or diff pair application. For a Vbe multiplier they may be OK,
Personally, I'm not tempted to try them
If you do try them, please let us know what you get?
By the way, regarding the KTC3200, I've never seen the so-called improved KTC3201 in a databook or made public outside of Japan.
Last edited:
Thank you for your great support.
I was pointing on this video BJT because i've seen them somewere as cascodes and VAS transistors. But you are right, normal build-for-audio 2sa1360 or sd1609 will outperform them in this applications.
I'll try to get some of the KTC3200 and check. Datasheet looks good
sanyo 2sa1016 seems to be same as sa992/sc1845, very good low current performance, higher noise for 2mA Ic operation
I was pointing on this video BJT because i've seen them somewere as cascodes and VAS transistors. But you are right, normal build-for-audio 2sa1360 or sd1609 will outperform them in this applications.
I'll try to get some of the KTC3200 and check. Datasheet looks good
sanyo 2sa1016 seems to be same as sa992/sc1845, very good low current performance, higher noise for 2mA Ic operation
I have seen this claim in other places. Has anybody ever published measured data to support this?
Onsemi datasheet contains some noise information
http://www.onsemi.com/pub_link/Collateral/2N4401-D.PDF
I did some noise measurements (mic preamp input stage) for 4401 vs 2240.
2240 was slightly better, especially at low frequencies. 4401 has quite high 1/f corner frequency.
But, still low enough for everything ex. maybe mic pre or MC head Input.
http://www.onsemi.com/pub_link/Collateral/2N4401-D.PDF
I did some noise measurements (mic preamp input stage) for 4401 vs 2240.
2240 was slightly better, especially at low frequencies. 4401 has quite high 1/f corner frequency.
But, still low enough for everything ex. maybe mic pre or MC head Input.
2SC2546-E Base Spreading Resistance
Hi, I was reading this thread and several others about how great the HITACHI/RENESAS 2SC2546/2SC2547 devices from a low-noise perspective.
I've had some experience with the HITACHI parts and better ROHM and TOSHIBA parts and was surprised to see posts (Syn08 et al) quoting Rbb as low as 1.9 Ohm for the 2SC2546-E - I believe their measurement technique is seriously flawed
My experience suggested that the Rbb for the 2SC2546/2SC2547 was a magnitude larger, around 11.8 ohms at 3mA (Ic).
Burkhard Vogel's book, The Sound Of Silence, 2nd edition, provides quite a nice analytical technique for calculating Rbb, where he shows that the value is around 13.5 Ohms, see page 57, however it doesn't take into consideration the Ic variation for the device under test; the approach tends to underestimate the actual Rbb.
My ex-colleague has actually measured Rbb for a range of Ic test cases and the results are published in the attached memo.
The attached memo is from a company I used to work for more than 10 years ago and has long since gone - I don't think anyone will mind if I post up the internal memo. I hope it will clear-up some of the misconceptions around the 2SC2546/2SC2547s Rbb value. (The 3rd page is lost in time - sorry)
At the time, I recall HITACHI not wanting to go public about their Rbb values as TOSHIBA and ROHM were going head-to-head on who had the lowest noise, single cell BJTs on the market. The ROHM and TOSHIBA devices were all single silicon junction devices, no Faulkner cells like the LM394 etc - there performance was pretty amazing according to a guy I knew from the old PHILIPS Semiconductor group in France.
Phil.
Hi, I was reading this thread and several others about how great the HITACHI/RENESAS 2SC2546/2SC2547 devices from a low-noise perspective.
I've had some experience with the HITACHI parts and better ROHM and TOSHIBA parts and was surprised to see posts (Syn08 et al) quoting Rbb as low as 1.9 Ohm for the 2SC2546-E - I believe their measurement technique is seriously flawed
My experience suggested that the Rbb for the 2SC2546/2SC2547 was a magnitude larger, around 11.8 ohms at 3mA (Ic).
Burkhard Vogel's book, The Sound Of Silence, 2nd edition, provides quite a nice analytical technique for calculating Rbb, where he shows that the value is around 13.5 Ohms, see page 57, however it doesn't take into consideration the Ic variation for the device under test; the approach tends to underestimate the actual Rbb.
My ex-colleague has actually measured Rbb for a range of Ic test cases and the results are published in the attached memo.
The attached memo is from a company I used to work for more than 10 years ago and has long since gone - I don't think anyone will mind if I post up the internal memo. I hope it will clear-up some of the misconceptions around the 2SC2546/2SC2547s Rbb value. (The 3rd page is lost in time - sorry)
At the time, I recall HITACHI not wanting to go public about their Rbb values as TOSHIBA and ROHM were going head-to-head on who had the lowest noise, single cell BJTs on the market. The ROHM and TOSHIBA devices were all single silicon junction devices, no Faulkner cells like the LM394 etc - there performance was pretty amazing according to a guy I knew from the old PHILIPS Semiconductor group in France.
Phil.
Last edited:
If you are looking for a very low noise bjt this is 2N4401/2N4403 combo.
Don't look in datasheet as will not tell you much, but this is the lowest base spreading resistance transistor on market.
And yes, will outperform 2SC2240 in terms of noise performance.
Regards,
Tibi
Keantoken did -
http://www.diyaudio.com/forums/anal...screte-opamp-open-design-329.html#post4157226
Reviving this old thread because it was very helpful to me two years ago when I needed to replace my 2SC458s to no avail since I could never find 2240s that were authentic. Recently, I got lucky. A local supplier had some hiding in their stock room and I just happened to stumble upon them. They had them recorded at $2.00 a piece from their 1980s inventory... I worked out a deal in exchange for some labor, and if anyone is looking for these hard to find authentic transistors, I've listed them on eBay:
Toshiba 2SC2240 GR 5 Pcs | eBay
Thank you
Toshiba 2SC2240 GR 5 Pcs | eBay
Thank you
Last edited:
Hi, I just noticed that both SC2240 and SA970 are now available from CDIL!
Reichelt.de has both in stock:
SC 2240: Transistor 2SC 2240 NPN 120V 0,1A TO-92 bei reichelt elektronik
- I wonder if they measure up to the originals?.. Will order some in any case
Reichelt.de has both in stock:
SC 2240: Transistor 2SC 2240 NPN 120V 0,1A TO-92 bei reichelt elektronik
- I wonder if they measure up to the originals?.. Will order some in any case
Yes 2N5551 will work I think, but observe that it has different pinout: EBC instead of ECB!