Makes sense. Why should LinearSystems turn the pinout 180 degrees for their
JFet which should be a direct replacement for the Toshiba SJ74.
But I remember that I spent a lot of time on the pinout of Toshiba 2SK170 /2SJ74 and the LinearSystems LSK170/LSJ74, when I was reading around here on DIYaudio. I remember my confusion when I was stuffing my BA-3 frontend boards and used Toshiba and LS devices.
Sorrry for the confusion! And tanks ZM for clarification!
I am climbing on the dumbness ladder
Greets
Dirk
JFet which should be a direct replacement for the Toshiba SJ74.
But I remember that I spent a lot of time on the pinout of Toshiba 2SK170 /2SJ74 and the LinearSystems LSK170/LSJ74, when I was reading around here on DIYaudio. I remember my confusion when I was stuffing my BA-3 frontend boards and used Toshiba and LS devices.
Sorrry for the confusion! And tanks ZM for clarification!
I am climbing on the dumbness ladder
Greets
Dirk
Dirk,
Just for clarification. Like I had said earlier. The pin outs for the LS and the Toshiba devices are the same. They ARE direct replacements. See updated pic for one more try. In post #388, you originally had your LS correct, and Toshiba incorrect. We've shown that the Toshiba is the same.
The confusion seems to be that one company (Toshiba) shows the device looking from the bottom on the datasheet, while LS shows the device from the top. See the link to the conversation I posted for more clarification re: the standards.
Most importantly, just absolutely know and be confident that they're the same pinouts and that your original drawing of the LS device is correct, and that mighty ZM has confirmed that the Toshiba is the same. And.... for just once... I got something correct.
Heck, I fall off the dumbness ladder quite often and go splat, so...
Cheers,
Patrick
Just for clarification. Like I had said earlier. The pin outs for the LS and the Toshiba devices are the same. They ARE direct replacements. See updated pic for one more try. In post #388, you originally had your LS correct, and Toshiba incorrect. We've shown that the Toshiba is the same.
The confusion seems to be that one company (Toshiba) shows the device looking from the bottom on the datasheet, while LS shows the device from the top. See the link to the conversation I posted for more clarification re: the standards.
Most importantly, just absolutely know and be confident that they're the same pinouts and that your original drawing of the LS device is correct, and that mighty ZM has confirmed that the Toshiba is the same. And.... for just once... I got something correct.
Heck, I fall off the dumbness ladder quite often and go splat, so...
Cheers,
Patrick
Here’s schematic of board I’m using: https://cdn.shopify.com/s/files/1/1006/5046/files/P-BAGS-1V20-schematic.pdf
I finished up my little preamp implementation of the BA1 FE. Thanks again ZM for your help!
I've attached the distortion measurement. I tested with several loads, and it was pretty consistent all the way across. Not surprisingly, sounds a lot like (and measures like) the ACP+. I used 2SK2013 for both of them, which could be part of that. It sounds fantastic right out of the box, but I may play with the degeneration and/or current source resistors to see if I can tweak spectra and THD in general. I put a few key resistors on sockets in anticipation. Honestly, I just love tweaking this stuff
Offset is stable and under control at 0mv, so also thinking of removing the 10uf coupling caps, which could be one advantage it has over the single supply ACP+.
All in all, a great front-end!
I've attached the distortion measurement. I tested with several loads, and it was pretty consistent all the way across. Not surprisingly, sounds a lot like (and measures like) the ACP+. I used 2SK2013 for both of them, which could be part of that. It sounds fantastic right out of the box, but I may play with the degeneration and/or current source resistors to see if I can tweak spectra and THD in general. I put a few key resistors on sockets in anticipation. Honestly, I just love tweaking this stuff
Offset is stable and under control at 0mv, so also thinking of removing the 10uf coupling caps, which could be one advantage it has over the single supply ACP+.
All in all, a great front-end!
Attachments
Last edited:
I successfully dropped the coupling caps on output today. This thing is sounding really good, especially paired with my F5. Has all the things I like from the ACP+, but also a nice bipolar, regulated supply (Salas UltraBIB) and less caps. I'd like to think I'm responding positively to the positive phase H2, but at .0036%, I'm not sure that's true
That a great question, and one I don’t have an answer for
Since I’m looping back the output of the BA1 instead of bootstrapped from an OS, I believe the gain would be reduced? Or maybe it’s just the voltage swing potential.
Conversely, with F5 I used some nice Toshiba outputs and found that I was playing with P3 to get quite a bit of distortion back into the mix. Instead, I figured I could just reduce the feedback level. I think I ended up going with the feedback scenario used in the F5 Turbo v1.
All of that to say that I probably get a little less than normal gain from my BA1 FE, and I get a little more than typical gain from F5. All in all, it’s plenty of gain for me.
I am amazed that there is not more builds and discussions of this amplifier. Having built most all of the FW clones including both the SE and PP Vfet amplifiers this is one of if not the best sounding SE I have heard. This is with a single output board per channel and not the full blown one using 2 output boards per channel. Careful attention to matching the output devices is strongly recommended. I bought sleeves of 25 where you will get same production run enabling matching easier. This is an older design but it seems most gravitate to newer ones overlooking one of the best builds for diyer's. Currently I am listening to mine with W Colburn's linestage which is also a great diy project. Together they are wonderful sounding with a soundstage worth hearing. Believe me you will not be disappointed if you like the SE sound. This amplifier along with the BA2 are the only amplifiers one needs to build for excellent sound. I used the BA3 front end for both my builds.
I agree with you. I did a variation of my own that sounded fantastic, two deep with Semisouth's. I could have easily lived happily ever after. But just moved on as new stuff came out to try. But many times those other attempts were a step backwards, looking back on it now. 25W of true single ended class A is no joke (PV1.4 = 1.26 X 104?? ). Just be ready for big hardware and a very warm listening room with 3.5 amps of bias current as intended. Some nice efficient 8 ohm speakers really helps to preserve the dynamics also, with this amp.
). Just be ready for big hardware and a very warm listening room with 3.5 amps of bias current as intended. Some nice efficient 8 ohm speakers really helps to preserve the dynamics also, with this amp.The BA-1 and BA-2 with the BA-3 front end are two of the best diy projects by NP. Most experienced amplifier builders know that multiple output transistors are a positive benefit with 3 pairs the sweet spot with the power factor of 25 watts. Just look at the insides of Passlabs amplifiers with the many, many output transistors. At least until the XA 25. The BA-2 output stage is essentially the F4 who many consider neutral sounding depending on the preamp used for flavor. I'll take my BA-1 with the BA-3 front end over the really good sounding SE V-fet every time. Nothing wrong with the SE V-fet it is just the BA-1 sounds much better to my ears and speakers.
Hi wdecho,
In his BA-3 Amplifier illustrated build guide, 6L6 introduces the 4 versions of the amplifier as follows:
- Differential FE, Complimentary output (BA-2)
- Differential FE, Single-ended output (BA-1)
- Complimentary FE, Complimentary output (BA-3)
- Complimentary FE, Single-ended output (BA-3 SE)
Thanks.
There are only two versions, PP and one SE but when initially was introduced they both had a different FE than the BA-3 front end. It gets somewhat confusing because most associate the BA-3 as the PP version with the BA-3 as the front end. But the BA-3 front end works just as well with the BA-1.