jdg123:
Wayne calls this a "line stage", not a preamp. The 100K input resistor to ground provides a load if an input pot is not used. Maximum input impedance would be 50K with a 100K pot installed. Probably no harm in removing R18 when using a permanently installed outboard pot, if you want a higher input impedance for some reason. I think the circuit sounds excellent with a 50K pot in parallel with R18. Keep in mind, with a higher input impedance, the noise will probably go up.
Wayne calls this a "line stage", not a preamp. The 100K input resistor to ground provides a load if an input pot is not used. Maximum input impedance would be 50K with a 100K pot installed. Probably no harm in removing R18 when using a permanently installed outboard pot, if you want a higher input impedance for some reason. I think the circuit sounds excellent with a 50K pot in parallel with R18. Keep in mind, with a higher input impedance, the noise will probably go up.
You da man!
Rick,
I first totally did not believe it could be the PSU—since it delivered stable values, and both channels got quite the same food [+18V GND -18V]...
Still , since I could, I swapped the VRDNs.
The Linestage has returned into the building...
It's floating, no connection to chassis, and each channel gets one rail's ground, and there's a complex noise [something between a fluttering hum, and a "rolling" highpitched, rhythmic "sinus". With or without denoiser doesn't matter.
Since nothing is where it should and all is surrounded with emitters [open wires, power amp underneath etc..., well], but will investigate anyway.
to all who already chimed in
a grand bow
and
thank you!
david
Rick,
I first totally did not believe it could be the PSU—since it delivered stable values, and both channels got quite the same food [+18V GND -18V]...
Still , since I could, I swapped the VRDNs.
The Linestage has returned into the building...
It's floating, no connection to chassis, and each channel gets one rail's ground, and there's a complex noise [something between a fluttering hum, and a "rolling" highpitched, rhythmic "sinus". With or without denoiser doesn't matter.
Since nothing is where it should and all is surrounded with emitters [open wires, power amp underneath etc..., well], but will investigate anyway.
to all who already chimed in
a grand bow
and
thank you!
david
Interesting.
On the standoffs...
I built my old Leach preamp with metal standoffs and star washers making solid continuity between board ground (it was laid in a ring trace around the edges) and the chassis.
But with the new projects I'm contemplating; a couple of Leach re-dos, a Pearl or two, and this linestage it seems more... strategic... to wire-solder grounding more particularly... more deliberately.
Employ ground planes sure but keep these separate from the chassis. Nylon. That's what I'm thinking.
Thoughts, guys?
On the standoffs...
I built my old Leach preamp with metal standoffs and star washers making solid continuity between board ground (it was laid in a ring trace around the edges) and the chassis.
But with the new projects I'm contemplating; a couple of Leach re-dos, a Pearl or two, and this linestage it seems more... strategic... to wire-solder grounding more particularly... more deliberately.
Employ ground planes sure but keep these separate from the chassis. Nylon. That's what I'm thinking.
Thoughts, guys?
Mark, since you chimed in (thank you 
)(and to all other knowledgeable fine folks here), allow me one of those pigeon-intelligent question:
Should the two rail's grounds be connected, or is the connection already established "inboard"?
While I'm at it, another one right away: I never thought that a standoff would actually make continuity between a board (ground) and the chassis... ???
edit: comment on chassis... the easy and reversible part seems to be a challenge to my ... character. The board's baseplate already looks like Verdun and it's not over yet...
For apologies, a little self-irony below...
)(and to all other knowledgeable fine folks here), allow me one of those pigeon-intelligent question:Should the two rail's grounds be connected, or is the connection already established "inboard"?
While I'm at it, another one right away: I never thought that a standoff would actually make continuity between a board (ground) and the chassis... ???
edit: comment on chassis... the easy and reversible part seems to be a challenge to my ... character. The board's baseplate already looks like Verdun and it's not over yet...
For apologies, a little self-irony below...
Attachments
Last edited:
@MLE that's a question I've been trying to get answered for the projects on my drafting board.
For this linestage and my original Leach, all the grounds are connected together on the boards (on the Leach each board had both channels). Your VRDN power supply may have separate grounds for +ve and -ve but these meet in any event at the GND terminal on the board... for each channel. Unless you're using dual supplies -- one for each channel these will "star" at the common PSU ground.
As I said the ground "rings" on the Leach boards were specifically intended to be connected to the chassis via the standoffs. I expect that the perimeter design was intended to function to an extent like a ground plane shield (but that's a guess). I've used my preamp for 40 years and it's worked pretty well.
On the boards I've been laying out I've established separate grounds (and ground planes) for each (+ve and -ve) on a separate board... separate from the signal circuitry... for each channel. I've done it this way because I'm powering these circuits with Super Regulators each of which obtains raw DC from an individual secondary unassociated from signal ground -- 4 secondaries in total. In this way I will have a lot of flexibility in establishing the order of grounds that works best for noise (shielding) and avoidance of loops. In that effort it seems best to keep the chassis separate from circuit grounds hence my initial suggestion to use insulating standoffs.
Well, that's the theory. I imagine this seems goofy as hell on this forum. I've asked a couple of times about aspects of this scheme and not heard back.
For this linestage and my original Leach, all the grounds are connected together on the boards (on the Leach each board had both channels). Your VRDN power supply may have separate grounds for +ve and -ve but these meet in any event at the GND terminal on the board... for each channel. Unless you're using dual supplies -- one for each channel these will "star" at the common PSU ground.
As I said the ground "rings" on the Leach boards were specifically intended to be connected to the chassis via the standoffs. I expect that the perimeter design was intended to function to an extent like a ground plane shield (but that's a guess). I've used my preamp for 40 years and it's worked pretty well.
On the boards I've been laying out I've established separate grounds (and ground planes) for each (+ve and -ve) on a separate board... separate from the signal circuitry... for each channel. I've done it this way because I'm powering these circuits with Super Regulators each of which obtains raw DC from an individual secondary unassociated from signal ground -- 4 secondaries in total. In this way I will have a lot of flexibility in establishing the order of grounds that works best for noise (shielding) and avoidance of loops. In that effort it seems best to keep the chassis separate from circuit grounds hence my initial suggestion to use insulating standoffs.
Well, that's the theory. I imagine this seems goofy as hell on this forum. I've asked a couple of times about aspects of this scheme and not heard back.
Last edited:
Rick,
I first totally did not believe it could be the PSU—since it delivered stable values, and both channels got quite the same food [+18V GND -18V]...
Still , since I could, I swapped the VRDNs.
The Linestage has returned into the building...
It's floating, no connection to chassis, and each channel gets one rail's ground, and there's a complex noise [something between a fluttering hum, and a "rolling" highpitched, rhythmic "sinus". With or without denoiser doesn't matter.
Since nothing is where it should and all is surrounded with emitters [open wires, power amp underneath etc..., well], but will investigate anyway.
to all who already chimed in
a grand bow
and
thank you!
david
Didn’t quite understand, is the problem solved or not?
If not, you might consider crushing the PCB with a sledgehammer. Should grt rid of the noise at least.
If not solved and you are looking for a solution involving still playing music, just chime in if you end up finding out the active devices are to blame. I’ll send.
Dennis always encourages me to use a scope to try and locate where in the circuit the noise appears. Might be worth a shot, though I never tried systematically myself.
For the record: Gnds are not connected between channels. If building balanced, Wayne said they should be connected. Or that you can experiment, like Mark said.
Cheers,
Andy
Last edited:
Hey Andy, thank you!
Hm, I'd say the situation has evolved.
The WLS is back, but it hums. I am quite convinced the board is intact ( )
Thank you for the offer, I'll get back to you if required.
I swapped the Outputstage (the amp), so I don't know if it's the same as before.
Scoping around is most probably one of the next requirements if I really want to tackle the problem(s)... bzzt-bzzt
Re: grounds, I already took some steps, like running it with variations of the grounds layout but should work a bit more ... seriously.
Hm, I'd say the situation has evolved.
The WLS is back, but it hums. I am quite convinced the board is intact (
) Thank you for the offer, I'll get back to you if required.I swapped the Outputstage (the amp), so I don't know if it's the same as before.
Scoping around is most probably one of the next requirements if I really want to tackle the problem(s)... bzzt-bzzt
Re: grounds, I already took some steps, like running it with variations of the grounds layout but should work a bit more ... seriously.
A question for Wayne, 6L6 or anyone else who has the answer:
The schematic calls for 2SK209BL (6.0 - 14 mA Idss) JFETS. If substituting 2SK209GR grade (2.6 - 6.5 mA Idss), should the 33 ohm source resistors (R4, R5) be jumpered to 0 ohms? Or possibly some other intermediate value? The LTSpice model shows:
BL Grade, 33 ohm source = 4.6 mA
GR grade, 33 ohm source = 3.0 mA
GR grade, 6 ohm source = 4.6 mA
GR grade, 0 ohm source = 5.3 mA
Is 5.3 mA still in the sweet spot for this JFET? 0 ohm jumpers are easy and seem to reduce the inherent DC offset to a few nano volts.
The schematic calls for 2SK209BL (6.0 - 14 mA Idss) JFETS. If substituting 2SK209GR grade (2.6 - 6.5 mA Idss), should the 33 ohm source resistors (R4, R5) be jumpered to 0 ohms? Or possibly some other intermediate value? The LTSpice model shows:
BL Grade, 33 ohm source = 4.6 mA
GR grade, 33 ohm source = 3.0 mA
GR grade, 6 ohm source = 4.6 mA
GR grade, 0 ohm source = 5.3 mA
Is 5.3 mA still in the sweet spot for this JFET? 0 ohm jumpers are easy and seem to reduce the inherent DC offset to a few nano volts.
I have a test PCB where I put in some BLs. Voltage drop across R1/2 increased significantly, to 43-ish mV (compared to Myleftears 20-ish mV drop using store parts kit, presumably GR). Drop across R7 and it’s partner across the board increased from 1V8 to 2V9, leaving about 19mA going to the LED instead of 12mA (might be wrong here, if so, sorry!).
This specific chan had some issues, but maybe worth of note, is that offset drift too seemed to increase. Not the equilibrium drift though, it just started at a higher point cold. With GRs, cold was approx 27mV, equilibrium 0. With Bl more like 50 at cold. Just an observation, like I said the circuit didn’t work as intended, others results may vary.
PS: 6L6 clarified yhe BL/GR misunderstanding some time in 2019 in the thread. In essence, schematic is wrong wrt grade. But Jim sez BL too works like a charm.
Regards,
Andy
This specific chan had some issues, but maybe worth of note, is that offset drift too seemed to increase. Not the equilibrium drift though, it just started at a higher point cold. With GRs, cold was approx 27mV, equilibrium 0. With Bl more like 50 at cold. Just an observation, like I said the circuit didn’t work as intended, others results may vary.
PS: 6L6 clarified yhe BL/GR misunderstanding some time in 2019 in the thread. In essence, schematic is wrong wrt grade. But Jim sez BL too works like a charm.
Regards,
Andy
Last edited:
A bit frustrating: DHL and the Diyaudiostore tell me my BA2018 kits were delivered yesterday but I received nothing
They might be in your post office. My replacements boards are, along with a customs bill to be paid on pickup.
I got a digital msg from my post service saying they have arrived. Expect a small letter to appear in my mailbox in a few days. They are slooooooow.
Last edited: