About Mosfets on external sinking surface prefer the chassis floor. Its enough for the 2x18 Ohm PSU CCS standard setting and you don't have to buy sinks. Don't bend them like that, lie them down along the board's edge facing up and bend the ends of their legs to pass underneath through the pads.
oops
Oh. Sorry. Didn't want to post bad ideas. I didn't do it myself, but there was someone's dcb1 like that. I thought it was a good thing because of the space gained.
Is it really not a good idea? Why? Too much stress on the legs?
Thank you for teaching me (/us)...
(should be more careful about jumping in, although I like the forum a lot)
david
Oh. Sorry. Didn't want to post bad ideas. I didn't do it myself, but there was someone's dcb1 like that. I thought it was a good thing because of the space gained.
Is it really not a good idea? Why? Too much stress on the legs?
Thank you for teaching me (/us)...
(should be more careful about jumping in, although I like the forum a lot)
david
For external sinking its better to first mount the MOSFETs on the sinking surface, then bend their legs to try how they best reach the board, soldering done last. So to avoid mechanical stress. Mezmerize is not designed for hard hot-rod hence the reservoir capacitors are what they are and the bridge diodes places are of small footprint. 2x18 Ohm setting is about right hot-rod for it. Most builders prefer to use the chassis as a sink for it. Because its dissipation is moderate. Flanking sinks are not excluded though.
Hypnotize was the legacy single input version that was provisioned for more hot-rod. That one was usually seen with large flanking sinks as a build. Had four snap-in reservoir caps, big CCS setting resistors, and along side TO-220 bridge diodes.
Hypnotize was the legacy single input version that was provisioned for more hot-rod. That one was usually seen with large flanking sinks as a build. Had four snap-in reservoir caps, big CCS setting resistors, and along side TO-220 bridge diodes.
I read the Threads about fake 2SK170's but still not certain if mine are fake or not. I got these a few years ago but certain where I bought them.
I'm measuring IDSS with a 10ohm resistor on the Drain with 9v, G&S shorted.
Will measuring IDSS determine if they are real or fake? So far one piece was wildly unstable and clearly bad, the others measure well and stable.
My first build I used pieces with a 9D date code as shown in the last pic. 20 pieces measured between 29 and 42 MV.
Any Ideas?
I'm measuring IDSS with a 10ohm resistor on the Drain with 9v, G&S shorted.
Will measuring IDSS determine if they are real or fake? So far one piece was wildly unstable and clearly bad, the others measure well and stable.
My first build I used pieces with a 9D date code as shown in the last pic. 20 pieces measured between 29 and 42 MV.
Any Ideas?
Attachments
Only in production replacement is Linear Systems LSK170B. The DIYA store has them in matched sets. Member myleftear recently suggested this source of unmatched & matched new old stock Toshiba 2SK170BL as genuine: Studio Electronics, INC.
Also a small stock of unmatched LSK170B right now appears at: NAC semi
All the above sources are located in the USA
Also a small stock of unmatched LSK170B right now appears at: NAC semi
All the above sources are located in the USA
Thanks, Salas
I just emailed LS directly for a quote for 100. Looking at the price from NAC I don't know if I'll be happy with the price quote, but I'll cross my fingers.
I just emailed LS directly for a quote for 100. Looking at the price from NAC I don't know if I'll be happy with the price quote, but I'll cross my fingers.
That's the same as Toshiba BL range (comes from blue, the Japanese pictured imaginary colors for IDSS hotness groups. Green, blue, violet). IDSS classification: GR: 2.6~6.5 mA, BL: 6.0~12 mA, V: 10~20 mA
Of course it isn't tight because its isn't about matched IDSS JFETs. Its about belonging to an IDSS group. For DCB1 you need 0.1mA tight matched quad you will handpick out of the BL (B) group by measurement. Those are the signal area units. For the rest of K170 positions in PSU area you may randomly pick remaining units out of their BL (Toshiba) or B (Linear Systems) groups.
Of course it isn't tight because its isn't about matched IDSS JFETs. Its about belonging to an IDSS group. For DCB1 you need 0.1mA tight matched quad you will handpick out of the BL (B) group by measurement. Those are the signal area units. For the rest of K170 positions in PSU area you may randomly pick remaining units out of their BL (Toshiba) or B (Linear Systems) groups.
https://www.ebay.com/p/1005633974
I believe these are original and not fake. Unless I'm not seeing things correctly!
Not to sure now.. But there are a few ebay sellers who state IDSS of 6-12ma for whatever it's worth.
I believe these are original and not fake. Unless I'm not seeing things correctly!
Not to sure now.. But there are a few ebay sellers who state IDSS of 6-12ma for whatever it's worth.
Last edited:
While wrong looking K170s are always fake we have also had a recent occasion of right looking ones that proved fake. Ebay parts is high risk taking. Especially the sought after NOS. Unless someone we know here has tested them in a DCB1 with correct results we can't foretell. See posts #2229 & #2248.
Will I obtain 4-6 matched quads, from 100 pieces of LSK170B's?
Can I use the low IDSS ones in DCB3?
Can I use the low IDSS ones in DCB3?
Last edited:
I don't remember that occasion. Very few members ever used 2SK170 in the DCG3 thread and they liked them but I think they were Toshiba. When I was developing DCG3 I had tried all possible JFETs and Linear Systems too. They measured close to K170 Toshiba. But bit better in HF THD trend and bit worse in THD total for the particular preamp because they have less parasitic capacitance and little less transconductance than the Toshibas. But it can possibly mean nothing in DCB1 where their mission is solo and much more simple. Not as voltage gain long tail pair and not part of an op-amp like circuit system.
Sounds to extreme to be even possible