GB "long and skinny" FirstWatt J2 clone PCBs

[Dr_Strangelove]

Always liked green the best. Green visually never bores me.

 

[mlwebb]

I like green.

 

[carsten.witt]

@Dennis,
I think for the EU members the same.
@All,
if it is ok for you I can take over EU Part of the GB but not in the next two/three weeks. I'm still to busy at work. Please let me know!

 

[thompsontechs]

Hello
please add me to the buy. i would like 1 set
thanks

North America & RoW
------------
Malvin 1 set
kevinheem 1 set
Soundslave 1 set
mattattnet 2 sets
Myint67 1set
twitchie 1 set
flamethrower1 1 set
necplusultra 1 set
WBS ???
Scott Stansbury - 1 set
nashbap 1set
mlwebb 1 set
Gnub 2 sets
Feds27 1 set (see page 21)
bologna1 1 set
leejh 1 set
AMB4 1 set
thompsontechs 2 sets



EU
------------
pfisla 2 sets
gig 1 set
Dr.Strangelove3 1 set
malisz ???
dieter12 2 sets
pistollero - 2 set
Spades 1 set
Rodeodave 1 set
MEPER 2 sets
iBuell 1 set


Unknown
------------
RustySanwa 2 sets

 

[Dennis Hui]

OK. Green boards, 35mM copper as Jeff's original then.

Since Carsten volunteered to handle things for the EU members, I will leave
that to him.

I will be ordering boards just for the other members.

Cheers,
Dennis

 

[Dennis Hui]

Hi thompsontechs,

Can you please confirm the number of sets you want?

Thanks,
Dennis

 

[carsten.witt]

@All,
as I wrote "... but not in the next two/three weeks .." if someone jump in please do it!

 

[thompsontechs]

Hi thompsontechs,

Can you please confirm the number of sets you want?

Thanks,
Dennis

Where we at on price Dennis?

 

[Dennis Hui]

About $25CAD for a pair including shipping (no tracking).

 

[thompsontechs]

About $25CAD for a pair including shipping (no tracking).

Confirm 2


Thanks

 

[Dennis Hui]

I should clarify that by 'pair' I meant a set of 2 PCBs, one left and one right channel.
So estimate is $25CAD for a stereo amp (1 left and 1 right channel PCB).

Dennis

 

[nashbap]

FYI, in my real J2 amp Papa used all bl grade jfets 2 each K170 and 2 each J174s per rail, and no dual 2jl109s. I have no way of measuring them though.

Just to be sure I understand that would be two J170's for the CCS(Q3 and Q4 in Jeff's schematic) and two J74's instead of the four J109's in the LTP(Q1A,Q1B,Q2A,Q2B)?

Am I correct in my understanding that the two sides of an LTP should match closely, so the Idss of Q1 should be equal to Q2? Also, if the CCS is producing the chosen 8.5ma(as per Jeff's analysis) do we need Q1 and Q2 to be greater than or equal to 8.5ma?

Thanks.
nash

 

[JeffYoung]

Hi Nash,

The SJ109s are a pair of SJ74s in a single plastic package. So you can use 2 SJ109s per channel (as the early factory J2s did), or 4 SJ74s (as hifibluedevil's factory J2 does).

You don't actually have to have the LTP doubled-up. You could use a single SJ109 per channel, or 2 SJ74s. But then it's not really a J2 -- or at least it's farther from the factory spec.

The two legs of the LTP should match. So in my schematic you'd match Q1A and Q1B, and then separately match Q2A and Q2B. It's not terribly important to match within a leg (ie: Q1A and Q2A).

The sum total of the LTP transistors should be well over 8.5mA. However, if you're using the factory spec of 4 BL-grade JFETS (whether 2 SJ109s or 4 SJ74s), then you're guaranteed to be have at least 24mA.

Cheers,
Jeff.

 

[Rodeodave]

I'm okay with waiting, no hurry.

 

[thompsontechs]

I should clarify that by 'pair' I meant a set of 2 PCBs, one left and one right channel.
So estimate is $25CAD for a stereo amp (1 left and 1 right channel PCB).

Dennis

Gotcha, make it 1 pair then.


PS, let me know where to paypal.

 

[nashbap]

Hi Nash,

The SJ109s are a pair of SJ74s in a single plastic package. So you can use 2 SJ109s per channel (as the early factory J2s did), or 4 SJ74s (as hifibluedevil's factory J2 does).

You don't actually have to have the LTP doubled-up. You could use a single SJ109 per channel, or 2 SJ74s. But then it's not really a J2 -- or at least it's farther from the factory spec.

The two legs of the LTP should match. So in my schematic you'd match Q1A and Q1B, and then separately match Q2A and Q2B. It's not terribly important to match within a leg (ie: Q1A and Q2A).

The sum total of the LTP transistors should be well over 8.5mA. However, if you're using the factory spec of 4 BL-grade JFETS (whether 2 SJ109s or 4 SJ74s), then you're guaranteed to be have at least 24mA.

Cheers,
Jeff.

Thanks Jeff for the detailed response on how you matched each leg of the LTP.


Regarding the ccs what does the factory J2 use?

The schematic shows R11 as 100k. I think your simulation used 110k or 120k for the 8.5ma result. Any change suggested?


nash

 

[JeffYoung]

Nelson told me factory J2s had CCS currents from 8mA to 12mA. (Or at least that's what I understood -- I hope I'm not misquoting.)

The simulations changed R11 just to be able to compare the results at the same gain (similar to volume matching when ABX testing). For a single instance the exact value isn't critical; I used 100K.

Cheers,
Jeff.

 

[nashbap]

Nelson told me factory J2s had CCS currents from 8mA to 12mA. (Or at least that's what I understood -- I hope I'm not misquoting.)
Jeff.

Thats why I cant understand how hifibluesdevil says he has two 170bl's in the CCS which at minimum would amount to 12ma; unless I am misreading him.
nash

 

[JeffYoung]

You're not misreading him. I've seen pictures.

(FWIW, I have measured BLs down to 5.9mA. But it still seems to be somewhat of a mystery.)

 

[Dr_Strangelove]

So later models switched from SJ109s to SJ74s.
The mystery continues...
Just sourced some genuine SJ109s.
I’ll just keep stashing parts for this one.