Calvin,
I hope that you'll recover, be well and make infinite number of good audio circuits in the future.🙂 We need you at these pages and respect your experience and suggestions.
Ivan
Hi,
thank You so much for all the good wishes. 🙂
Still though I´ll withdraw from this thread, as I think all important info is given.
There were circuit design and part suggestions developing the initial idea of a specialized cute small buffer of #1 to a more common, groupable design.
After #16 there was imho only the Q left, wether to TH or to SMD the thingie.
Should have been easy to answer, since the suggested active parts are available in both styles.
Instead, from #30 on the thread went straight down the pipe, beeing now Reset-to-Start with #80.
We´ll sure meet in other threads.
Till then...
jauu
Calvin
thank You so much for all the good wishes. 🙂
Still though I´ll withdraw from this thread, as I think all important info is given.
There were circuit design and part suggestions developing the initial idea of a specialized cute small buffer of #1 to a more common, groupable design.
After #16 there was imho only the Q left, wether to TH or to SMD the thingie.
Should have been easy to answer, since the suggested active parts are available in both styles.
Instead, from #30 on the thread went straight down the pipe, beeing now Reset-to-Start with #80.
We´ll sure meet in other threads.
Till then...
jauu
Calvin
Isn't this essentially a Pass B1 type circuit with a BJT current source instead of the FET?
/Martin
/Martin
I've done a bit more sim work on this - attached are the 4 circuits I've evaluated. I tried a number of different types, and the input capacitance has a big impact on distortion - some of the devices with 25 to 30pF produce 0.01% at 20 kHz (its not Idss related - I checked that - its pure input capa that's causing it).
Cascoding helps, but if you use a low input capacitance JFET, the additional benefit is small - look at the results for the LHS buffer using a low input capacitance device.
If we went for a BF862 RF JFET, we'd get a cheap, readily available JFET with low input capacitance.
But, its SMD . . .
Cascoding helps, but if you use a low input capacitance JFET, the additional benefit is small - look at the results for the LHS buffer using a low input capacitance device.
If we went for a BF862 RF JFET, we'd get a cheap, readily available JFET with low input capacitance.
But, its SMD . . .
We should be using SMD because that's where the greatest flexibility in parts choice is. Making an inferior design just to stay with pin in hole devices isn't of interest to me. I'm over 60 years old and have no issue placing and soldering SMD parts. If I can do it anyone can.
people should use what they like/need, but for the right reasons ie. best performance, best price, easiest availability. I dont want to force anyone to do anything, I may build this someday, but not any time soon. my purpose for posting was merely to provide a bit of a reality check and remind people of the changing world we as DIYers MUST come to terms with if we wish to continue the hobby and that reality is here now, especially for those just starting the hobby who do not have a stash of NOS parts. I of course believe in all manner of other reasons SMD is superior anyway, but i've already painted that picture.
Last edited:
OK. Let's use SMD j-fet. I am only affraid that the pcb will become less attractive for some (many?) potential group buyers. I suppose that with this decision the whole pcb must use SMD and therefore simpler circuit would be most welcome.
I am looking for a decent easily available TH JFET with low input capacitance for a TH version. I can do two boards no problem.
LSK489?, although not that readily avail, does come in 3 diff pkgs, TH TO-72. Lower Ciss (4p) than BF862(10p). If folks are having issues with SMT, try to standardize on 1206 pkg for passives.
I am not that much of a RF designer, but I guess Crss & Ciss determine Ft?
I am not that much of a RF designer, but I guess Crss & Ciss determine Ft?
LSK device looks very good indeed but they are difficult to get hold of. BF244B also a nice device for this application and I see Mouser have a few thousand of the TO-92 in stock.
There ya go THT folks, BF244B End of Life: one by one they go ...
I agree, BF862 is the top performer in so many ways.
I agree, BF862 is the top performer in so many ways.
Yes, but LSK489 is around 6.- dollars each if you buy one thousand. Even if we find 100 group buyers and buy one LSK489 for them all it would be considerable investment for the GB organizer! And buying them individually is almost mission impossible.
LSK489, Mission impossible, not really,
I bought some LSK389B from the dealer in Florida, called, gave VISA#, sent, all about 15 minutes of work. I do agree that it would be nice to get all parts from 1 (preferred) or 2 suppliers only , Newark/Mouser/Digi-Key, as shipping costs are high. For Mouser/Digi-Key I try to get it over $200, to get free shipping.
Built the Elektor 500ppm LCR meter, had to get one switch from RS components = crazy. Soaked me with DHL costs. Parts made from the junk box or obscure suppliers is a PITA.
I bought some LSK389B from the dealer in Florida, called, gave VISA#, sent, all about 15 minutes of work. I do agree that it would be nice to get all parts from 1 (preferred) or 2 suppliers only , Newark/Mouser/Digi-Key, as shipping costs are high. For Mouser/Digi-Key I try to get it over $200, to get free shipping.
Built the Elektor 500ppm LCR meter, had to get one switch from RS components = crazy. Soaked me with DHL costs. Parts made from the junk box or obscure suppliers is a PITA.
And let's settle for BF244 for the TH board, there is still 6 thousands of them available from Mouser.
Is it possible to incorporate the sot23 pad layout for an SMD inside or alongside the pad layout of the through hole To92 device?
A GDS (or SDG) To92 pin out is the most common for jFETs. Only the the 2N devices gave the Source as the middle pin.
A sot23 is 1=S, 2=D, 3=G and pin3 is the one on the other side of the package. Surely we just rotate the smd 130degrees to align the pads of both packages.
If this is possible (it does not create any additional PCB area) then we can incorporate both SMD & To92 pads for every transistor location.
A GDS (or SDG) To92 pin out is the most common for jFETs. Only the the 2N devices gave the Source as the middle pin.
A sot23 is 1=S, 2=D, 3=G and pin3 is the one on the other side of the package. Surely we just rotate the smd 130degrees to align the pads of both packages.
If this is possible (it does not create any additional PCB area) then we can incorporate both SMD & To92 pads for every transistor location.
It's easy to include two sets of holes for the rear pot (of stereo pot) distanced 5 mm. In this way both ALPS Blue Velvet RK27 and RK16 could be used (and switched attenuator DACT21 as suggested by JP).
Here is the fist cut SMD board. This board is laid out for the stereo RK27 pot.
TH version will have to wait for a few weeks - I have to do some more work on my big preamp first.
I am still cleaning up this layout - thickening tracks up, opening up clearances etc, but it should give some idea.
Anyway, take a look. BTW, note, this design requires JFETS with low input capacitance and Idss of at least 8mA - so BF862 or BF244 - any other sugestions please let me know and I will try to accomodate in the layout.
TH version will have to wait for a few weeks - I have to do some more work on my big preamp first.
I am still cleaning up this layout - thickening tracks up, opening up clearances etc, but it should give some idea.
Anyway, take a look. BTW, note, this design requires JFETS with low input capacitance and Idss of at least 8mA - so BF862 or BF244 - any other sugestions please let me know and I will try to accomodate in the layout.
Last edited:
- Status
- Not open for further replies.