Wow-- this is what had me on the edge for actually doing the lxmini's..
Thanks, and can't wait to see them on the storefront.
Thanks, and can't wait to see them on the storefront.
Free giveaway for LXmini builders
Hi,
we have parts from our LXmini-kit production as a free giveaway:
LXmini construction plans (outside EU 19% VAT is deducted after registering).
Shipping is free within Germany,
20€ for EU , 40€ outside EU
Contact info@magicLX521.com with your shipping address.
Best
Frank
Hi,
we have parts from our LXmini-kit production as a free giveaway:
- Round LXmini baseplates, HDF precision mitered, dark anthracite with lower-side cable canals and 5 predrilled mounting holes for rubber feet
- 4" end caps for the lower tubes
LXmini construction plans (outside EU 19% VAT is deducted after registering).
Shipping is free within Germany,
20€ for EU , 40€ outside EU
Contact info@magicLX521.com with your shipping address.
Best
Frank
Good news and thank you papa for another great work.
I am considering building +2 or studio, should I wait for the next release or I can build these boards and just add on the subwoofer channel like an "expansion" board?
Example in balance mood, 2 of these PCB + 1 or 2 future PCB for the sub?
I am considering building +2 or studio, should I wait for the next release or I can build these boards and just add on the subwoofer channel like an "expansion" board?
Example in balance mood, 2 of these PCB + 1 or 2 future PCB for the sub?
I am subscribed to this thread, so it isn't that much trouble to wait, but is there a timeline for putting the kits on the storefront? Just want to order the day it comes out 🙂
It will keep me from checking a non-updated store front 2x daily.
Thanks,
It will keep me from checking a non-updated store front 2x daily.
Thanks,
Last edited:
Will the LXMini+2 and LXMini Studio be a completely different PCB/kit, or an add-on to the LXMini ASP?
They really are an amazing set of speakers...can’t wait to pair them with the ASP.
They really are an amazing set of speakers...can’t wait to pair them with the ASP.
Nelson,
I hope you don't mind a few questions / remarks.
I find it a pity that you spend much of your article on the NJFET follower, and little or nothing on the Sallen Keys.
Undoubtedly using a pair of J113 to get such low distortion is very clever.
But it is perhaps worth mentioning that in doing so, the DC across input and output of the follower is not zero.
It is only made possible by using single rail and decoupling caps.
So split rail and DC couple would not work, if I am not wrong.
In any case I find that this is still a compromise.
What made you choose J113 when you can choose 2SK209GR instead, which is equally low cost and available ?
And IMHO better than J113 in many respect.
You also use a somewhat heavy gate stopper for the signal carrying FETs, but nothing for the CCS's.
Is that a random choice, or did you experience oscillations ?
1k is not a small value when some of the resistors are even below 10k in the Sallen Keys.
Perhaps you would also kindly tell us how you chose the parameters for the Sallen Keys (e.g. what Q value).
And whehter you decided to swap the first and second stages from your original schematics (P.2) to the final (?) one in P.7 for specific reasons.
Thanks in advance,
Patrick
I hope you don't mind a few questions / remarks.
I find it a pity that you spend much of your article on the NJFET follower, and little or nothing on the Sallen Keys.
Undoubtedly using a pair of J113 to get such low distortion is very clever.
But it is perhaps worth mentioning that in doing so, the DC across input and output of the follower is not zero.
It is only made possible by using single rail and decoupling caps.
So split rail and DC couple would not work, if I am not wrong.
In any case I find that this is still a compromise.
What made you choose J113 when you can choose 2SK209GR instead, which is equally low cost and available ?
And IMHO better than J113 in many respect.
You also use a somewhat heavy gate stopper for the signal carrying FETs, but nothing for the CCS's.
Is that a random choice, or did you experience oscillations ?
1k is not a small value when some of the resistors are even below 10k in the Sallen Keys.
Perhaps you would also kindly tell us how you chose the parameters for the Sallen Keys (e.g. what Q value).
And whehter you decided to swap the first and second stages from your original schematics (P.2) to the final (?) one in P.7 for specific reasons.
Thanks in advance,
Patrick
I measure 0.003% with j113, idss 18ma, Iq 7ma, Rs 100 ohm.
No caps, power supply +9.3 and - 12V.
Only third harmonic.
1vrms out at 1khz at 4k load.
Better then untweaked toshiba.
No caps, power supply +9.3 and - 12V.
Only third harmonic.
1vrms out at 1khz at 4k load.
Better then untweaked toshiba.
Yes, using a very high Idss and then include a large source resistor will make things look much better.
I never need to tweak Toshiba's.
And then we can also start to compare Zout, noise, max. voltage, ..., etc.
But we are both happy with what we use, so everyone is happy. 😉
Cheers,
Patrick
I never need to tweak Toshiba's.
And then we can also start to compare Zout, noise, max. voltage, ..., etc.
But we are both happy with what we use, so everyone is happy. 😉
Cheers,
Patrick
I find it a pity that you spend much of your article on the NJFET follower, and little or nothing on the Sallen Keys.
It is only made possible by using single rail and decoupling caps.
So split rail and DC couple would not work, if I am not wrong.
What made you choose J113 when you can choose 2SK209GR instead, which is equally low cost and available ?
And IMHO better than J113 in many respect.
You also use a somewhat heavy gate stopper for the signal carrying FETs, but nothing for the CCS's.
Is that a random choice, or did you experience oscillations ?
1k is not a small value when some of the resistors are even below 10k in the Sallen Keys.
Perhaps you would also kindly tell us how you chose the parameters for the Sallen Keys (e.g. what Q value).
And whehter you decided to swap the first and second stages from your original schematics (P.2) to the final (?)
one in P.7 for specific reasons.
In order of asking,
The filter parameters were the ones approved by SL after listening
tests. Sallen Key design is old hat - much more interested in fiddling
way to use cheap/available fets.
The project is intended for cheap single supply wall-wart switcher. Split
supplies are another category of cost and availability. DC coupled version
is a whole other category of DIY complexity and cost.
I have nice artwork for the same using the 2SK170/2SJ74 if people
want to pay for it...
2SK209? I go to Digikey and see discontinued which tells me not
to bother specifying in future projects. Also it's surface mount, a real
treat for me and most DIYers.
1K is a heavy gate stopper on the Jfet? It's been my default for years,
even on the Beast with 1,000 Jfets.
The CCS doesn't seem to need it, so I left it off, and that has also worked
for years without problems.

Thanks Nelson.
I just checked 2SK209 at Mouser and Digikey.
Mouser has 11k+ in stock.
DK has 3k+ in stock, part no. 2SK209-GR(TE85LF)CT-ND
And the Toshiba website did not mention it is EOL.
Cheers,
Patrick
I just checked 2SK209 at Mouser and Digikey.
Mouser has 11k+ in stock.
DK has 3k+ in stock, part no. 2SK209-GR(TE85LF)CT-ND
And the Toshiba website did not mention it is EOL.
Cheers,
Patrick
It's an OK part.
Buy them while you can - I say the writing's on the wall. It says
"discontinued at Digikey" on the listing and Mouser's 11K is "on order"
with a 10 week quote.

Buy them while you can - I say the writing's on the wall. It says
"discontinued at Digikey" on the listing and Mouser's 11K is "on order"
with a 10 week quote.

Only a meager selection of Sanyo/On Semi low noise n channel left but they are all 15V Vdsx parts.
The 2SK3666 may be a bit more useful at 30V Vdsx and Yfs 6.5 mS, but has 2x noise which may still be passable for line level duty.
The 2SK3666 may be a bit more useful at 30V Vdsx and Yfs 6.5 mS, but has 2x noise which may still be passable for line level duty.
It's an OK part.
Apparently someone in Pass Labs use them in reels for quite some years.
http://www.diyaudio.com/forums/analog-line-level/218373-discrete-opamp-design-21.html#post3138516
Can't be that bad. 🙂
PS 2SK2145 = Dual 2SK209
Cheers,
Patrick
- Home
- Amplifiers
- Pass Labs
- LX-mini Crossover Article