Re: 6s19pi
It's breadboarded, but some of the parts for the CCS were held by my supplier because others were on backorder. I've heard it breifly RC loaded, but because the B+ is too low for that, it was far from optimal. I also got distracted with the PP2C and my 813PP (only awaiting PSU caps - aaaargh!). Been quite ill too.
I will let you know soon. Unless there is something totally wrong with the tube, I've a feeling it might be a killer.
Chhers
Hi Bas,Bas Horneman said:
It's breadboarded, but some of the parts for the CCS were held by my supplier because others were on backorder. I've heard it breifly RC loaded, but because the B+ is too low for that, it was far from optimal. I also got distracted with the PP2C and my 813PP (only awaiting PSU caps - aaaargh!). Been quite ill too.
I will let you know soon. Unless there is something totally wrong with the tube, I've a feeling it might be a killer.
Chhers
Hi argo,
Thanks for your schematic. I think I will try it, then "report" my experiences about it.
Regards, Miki.
Ps. I started to build an SRPP, with 184V voltage, two 330 ohm resistors (one betwen the tubes, one cathod resistor), and I will try to tune it for my needs. I will post it if it works, before I don't want to post a posibbly wrong sch. The calculated parameters are Ok. ~800 Ohm output impedance, 25x gain.
Thanks for your schematic. I think I will try it, then "report" my experiences about it.
Regards, Miki.
Ps. I started to build an SRPP, with 184V voltage, two 330 ohm resistors (one betwen the tubes, one cathod resistor), and I will try to tune it for my needs. I will post it if it works, before I don't want to post a posibbly wrong sch. The calculated parameters are Ok. ~800 Ohm output impedance, 25x gain.
this is a reply to argo's feature pic of 6922/6dj8
first the voltage of B+ is to low
with plate resistor of 10k I=150/10000 the plate current of 15ma
i feel that's a bit high plus with a low B+ the voltage swing is low
after the first stage to second stage u are directly coupling it to the next tube. with out a coupling cap install there would be + voltage on the grid of the second tube. Hence the second tube would not conduct unless the cathode is applied with more +voltage then the grid voltage due to first stage
the grid of the second stage is floating also try using a resistor to ground to fix the voltage of the grid to zero use a 1M resistor or little bit less would do so much harm.
your cathode follower is wired wrongly. One of the proper way can be seen in the first page of this post(6sn7)
for a single stage triode driving a cathode follower the effect would be inverting so do the nessecary adjustments.
also for single stage triode without a ccs. the tube would not be very linear(compared to a tube with ccs). hence use a ccsource at the first tube to linearize the first tube. Also cathode follower are better built around ccsinks makes them more linear. Don't expect the tube linearize itself by chosen the best load line.
The tube chosen would be a go choice of tube. Has hence a low rp. hence a better cathode follower.
continue posting your update maybe we can get some where with this design. i will try to calculate the resistor to check for the right value. I was told that for 6922 anode follower(common cathode) the plate voltage should be around 90v and the current should be 2.2ma( I was told). Another rule of thumb is to design the total plate dissipation to 1/4 of the total dissipation.
good luck modding the circuit then
first the voltage of B+ is to low
with plate resistor of 10k I=150/10000 the plate current of 15ma
i feel that's a bit high plus with a low B+ the voltage swing is low
after the first stage to second stage u are directly coupling it to the next tube. with out a coupling cap install there would be + voltage on the grid of the second tube. Hence the second tube would not conduct unless the cathode is applied with more +voltage then the grid voltage due to first stage
the grid of the second stage is floating also try using a resistor to ground to fix the voltage of the grid to zero use a 1M resistor or little bit less would do so much harm.
your cathode follower is wired wrongly. One of the proper way can be seen in the first page of this post(6sn7)
for a single stage triode driving a cathode follower the effect would be inverting so do the nessecary adjustments.
also for single stage triode without a ccs. the tube would not be very linear(compared to a tube with ccs). hence use a ccsource at the first tube to linearize the first tube. Also cathode follower are better built around ccsinks makes them more linear. Don't expect the tube linearize itself by chosen the best load line.
The tube chosen would be a go choice of tube. Has hence a low rp. hence a better cathode follower.
continue posting your update maybe we can get some where with this design. i will try to calculate the resistor to check for the right value. I was told that for 6922 anode follower(common cathode) the plate voltage should be around 90v and the current should be 2.2ma( I was told). Another rule of thumb is to design the total plate dissipation to 1/4 of the total dissipation.
good luck modding the circuit then
Nick said,
<b>first the voltage of B+ is to low</b>
Agreed. It needs to be juggled up a bit to about Va = 100V to make the direct coupling work better.
<b>with plate resistor of 10k I=150/10000 the plate current of 15ma</b>
No. The way you've calculated it doesn't factor the voltage across the tube or Rk. You have the whole B+ across Ra. Apart from that 12 - 15mA is about where the 6DJ8 sounds best in my experience resistively loaded (20mA average limiting cathode current)
Plugging it all into Tubecad, I got,
Va = 77V
Ia = 7.23mA
Vg = -1.6V
Pa = 558mW
<b>after the first stage to second stage u are directly coupling it to the next tube. with out a coupling cap install there would be + voltage on the grid of the second tube. Hence the second tube would not conduct unless the cathode is applied with more +voltage then the grid voltage due to first stage</b>
Yep, but why not just juggle the parameters of the CF to get the cathode more +ve than the grid? Better than adding a cap in the signal path. Caveat: DC coupling is not a beginner's sport and I would actually recommend argo add the RC network for safety with the selected DC op points and remove them when the circuit is shown to work.
<b>also for single stage triode without a ccs. the tube would not be very linear(compared to a tube with ccs). hence use a ccsource at the first tube to linearize the first tube. Also cathode follower are better built around ccsinks makes them more linear. Don't expect the tube linearize itself by chosen the best load line.</b>
Agreed, but even with a CCS, the 6DJ8 family aren't all that linear.
<b>the current should be 2.2ma( I was told). Another rule of thumb is to design the total plate dissipation to 1/4 of the total dissipation.</b>
Disagree, the 6DJ8 sounds awful at low currents and it's linearity is poor and Rp high. Lots of current is better.
Pa = 80% of rating is quite a sound tradeoff for life/performance.
Va = 100V, Ia = 12mA means Pa = 1.2W is only 2/3 of it's rating.
All my comments in previous threads about CF's stand, but if you <i>must</i> use one the line stage of <a href="http://www.vacuumstate.com/images/FVP5_Rev_1.GIF">this</a> is much better. It measures exceptionally well, and is the best sounding CF I've ever heard, drives reactive loads well, has a gain of about 15 (still too much for most linestages) and a Zout of ~ 200R.
<b>first the voltage of B+ is to low</b>
Agreed. It needs to be juggled up a bit to about Va = 100V to make the direct coupling work better.
<b>with plate resistor of 10k I=150/10000 the plate current of 15ma</b>
No. The way you've calculated it doesn't factor the voltage across the tube or Rk. You have the whole B+ across Ra. Apart from that 12 - 15mA is about where the 6DJ8 sounds best in my experience resistively loaded (20mA average limiting cathode current)
Plugging it all into Tubecad, I got,
Va = 77V
Ia = 7.23mA
Vg = -1.6V
Pa = 558mW
<b>after the first stage to second stage u are directly coupling it to the next tube. with out a coupling cap install there would be + voltage on the grid of the second tube. Hence the second tube would not conduct unless the cathode is applied with more +voltage then the grid voltage due to first stage</b>
Yep, but why not just juggle the parameters of the CF to get the cathode more +ve than the grid? Better than adding a cap in the signal path. Caveat: DC coupling is not a beginner's sport and I would actually recommend argo add the RC network for safety with the selected DC op points and remove them when the circuit is shown to work.
<b>also for single stage triode without a ccs. the tube would not be very linear(compared to a tube with ccs). hence use a ccsource at the first tube to linearize the first tube. Also cathode follower are better built around ccsinks makes them more linear. Don't expect the tube linearize itself by chosen the best load line.</b>
Agreed, but even with a CCS, the 6DJ8 family aren't all that linear.
<b>the current should be 2.2ma( I was told). Another rule of thumb is to design the total plate dissipation to 1/4 of the total dissipation.</b>
Disagree, the 6DJ8 sounds awful at low currents and it's linearity is poor and Rp high. Lots of current is better.
Pa = 80% of rating is quite a sound tradeoff for life/performance.
Va = 100V, Ia = 12mA means Pa = 1.2W is only 2/3 of it's rating.
All my comments in previous threads about CF's stand, but if you <i>must</i> use one the line stage of <a href="http://www.vacuumstate.com/images/FVP5_Rev_1.GIF">this</a> is much better. It measures exceptionally well, and is the best sounding CF I've ever heard, drives reactive loads well, has a gain of about 15 (still too much for most linestages) and a Zout of ~ 200R.
Oh looky, there are two 6DJ8 sections in the circuit. The FETS are used in an appropriate way as helpers for the tube. Tubes can be subbed for the FETs but will require a higher B+ and probably won't work as well (unless pentodes).Joel said:
That's funny, I thought this was the tubes forum.
Why the "mad" icon? . You attempting another cheap shot?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.