Hi
I just got 1000H gridchoke and I intend use it in my 300b Waltons/Jelab Circuit with single 6sn7 tube.
I allready connected Choke parallel with old 220k grid res to ground but that doesnt seem to be right? Output bias increased from -66 to near 100 vdc . Choke seems to have 24k DC res.
How to use it?? Thanks for help.
Papparazzi
I just got 1000H gridchoke and I intend use it in my 300b Waltons/Jelab Circuit with single 6sn7 tube.
I allready connected Choke parallel with old 220k grid res to ground but that doesnt seem to be right? Output bias increased from -66 to near 100 vdc . Choke seems to have 24k DC res.
How to use it?? Thanks for help.
Papparazzi
Hi,
I suspect your 300B tubes are perhaps a bit gassy and with a 220K grid resistor the grid current was generating some additional contact potential bias.
Some grid current in 300B is not unusual, but I think more than ~5 - 10 uA should be a cause for concern. Whose 300B are you using? I only ask because in the past I have encountered unusually high grid currents in both Shuguang and Valve Arts 300B tubes operated at 400V.
IMO 100K should be the upper limit, particularly if fixed bias is used.
The DCR of the grid choke is much lower. To test this hypothesis, put a 200K resistor in series with the choke, if the cathode current then drops back to about where it was the hypothesis is confirmed.
The solution is simple, just increase the value of the cathode bias resistor until the operating current is the same. No big deal.
The grid choke is commonly used here and I' m surprised you have been unable to find schematics showing this.
When you use a grid choke you also end up with a subsonic high q resonance that can be most easily fixed (if you don't like the way it sounds
) by increasing the value of the coupling capacitor. There are ways of reducing the Q of the resonance, but entail adding varying amounts of R in series/parallel with the choke. Incidentally adding parallel R sort of defeats the point of a grid choke in most cases. 
Edits: I fixed a lot of typos, apparently I can't think and type at the same time..
I suspect your 300B tubes are perhaps a bit gassy and with a 220K grid resistor the grid current was generating some additional contact potential bias.
Some grid current in 300B is not unusual, but I think more than ~5 - 10 uA should be a cause for concern. Whose 300B are you using? I only ask because in the past I have encountered unusually high grid currents in both Shuguang and Valve Arts 300B tubes operated at 400V.
IMO 100K should be the upper limit, particularly if fixed bias is used.
The DCR of the grid choke is much lower. To test this hypothesis, put a 200K resistor in series with the choke, if the cathode current then drops back to about where it was the hypothesis is confirmed.
The solution is simple, just increase the value of the cathode bias resistor until the operating current is the same. No big deal.
The grid choke is commonly used here and I' m surprised you have been unable to find schematics showing this.
When you use a grid choke you also end up with a subsonic high q resonance that can be most easily fixed (if you don't like the way it sounds
) by increasing the value of the coupling capacitor. There are ways of reducing the Q of the resonance, but entail adding varying amounts of R in series/parallel with the choke. Incidentally adding parallel R sort of defeats the point of a grid choke in most cases. Edits: I fixed a lot of typos, apparently I can't think and type at the same time..
Hi Papparazzi,
Hmm, I have heard a lot of positive press on the TJ, but it does sound like there is really a lot of grid current with these particular tubes or there is something else going on. (Like a faulty meter - albeit unlikely as that is or HF/VHF oscillations.)
I would insert the original 220K resistors in series with the chokes and remeasure the the voltage across the bias resistors, I suspect it will be right around 64 - 66V.
If you have a dvm with a sufficiently high input resistance (say 10M) you might be able to measure the voltage at the grid. I have never encountered a grid current induced voltage drop of more than a couple of volts with any half way good 300B.. YMMV though..
Your current circuit implementation if you have the choke connected where the old 220K resistor was is correct. You do have a 1K grid stopper resistor in series right at the grid of the 300B don't you?
(I'm thinking HF/VHF oscillation here, which you probably cannot measure with the instruments at hand.) Try adding stoppers if not already present and then remeasure the cathode bias voltage.
If all this checks out ok, and bias is still too high an acceptable solution for the TJ anyway would be to change the 1K cathode resistor to 1.5K, this should put things back exactly where they were.
Hmm, I have heard a lot of positive press on the TJ, but it does sound like there is really a lot of grid current with these particular tubes or there is something else going on. (Like a faulty meter - albeit unlikely as that is or HF/VHF oscillations.)
I would insert the original 220K resistors in series with the chokes and remeasure the the voltage across the bias resistors, I suspect it will be right around 64 - 66V.
If you have a dvm with a sufficiently high input resistance (say 10M) you might be able to measure the voltage at the grid. I have never encountered a grid current induced voltage drop of more than a couple of volts with any half way good 300B.. YMMV though..
Your current circuit implementation if you have the choke connected where the old 220K resistor was is correct. You do have a 1K grid stopper resistor in series right at the grid of the 300B don't you?
(I'm thinking HF/VHF oscillation here, which you probably cannot measure with the instruments at hand.) Try adding stoppers if not already present and then remeasure the cathode bias voltage.
If all this checks out ok, and bias is still too high an acceptable solution for the TJ anyway would be to change the 1K cathode resistor to 1.5K, this should put things back exactly where they were.
I have 1k grid stopper .Using Grid choke to to ground also drops B+ voltage about 35 volts according to PSU impedance .I might solve the problem increasing 1k to 1,5k ,but does these tubes anymore work correctly ,what is the real current drawn through Cathode? Is the only way to measure current through primary DC res.
papparazzi
papparazzi said:
Maybe this will help - http://www10.brinkster.com/azcruz/My5010schema.html
For what ever it's worth, I've tried the TJ globes on this circuit.
All of the current in the tube also flows through the cathode circuit, it is perfectly valid to measure the voltage across a known resistance and calculate the current from that.
I'm not surprised that the plate voltage has sagged by 35V, the increase in plate current is equivalent to about a 50% increase in overall load current, and undoubtedly the supply was not designed to handle this much additional current.
Depending on rectifier tube and transformer you might not want to run the amplifier until you replace the cathode resistors with 1.5K ones.
Replacing the resistors with 1.5K ones will result in the same operating point as you had before the installation of the grid chokes.
I am still concerned about the large implied grid current of your output tubes - do you have some non TJ 300B you can use to try to establish a baseline? (See if this is true of other 300B as well?)
IMO the only cathode biased 300B amp I have done had 1.5K cathode resistors and a raw plate voltage of 400V, grid resistors were 50K IIRC.. Idle current was about 55mA.. Tubes were Valve Arts..
I'm not surprised that the plate voltage has sagged by 35V, the increase in plate current is equivalent to about a 50% increase in overall load current, and undoubtedly the supply was not designed to handle this much additional current.
Depending on rectifier tube and transformer you might not want to run the amplifier until you replace the cathode resistors with 1.5K ones.
Replacing the resistors with 1.5K ones will result in the same operating point as you had before the installation of the grid chokes.
I am still concerned about the large implied grid current of your output tubes - do you have some non TJ 300B you can use to try to establish a baseline? (See if this is true of other 300B as well?)
IMO the only cathode biased 300B amp I have done had 1.5K cathode resistors and a raw plate voltage of 400V, grid resistors were 50K IIRC.. Idle current was about 55mA.. Tubes were Valve Arts..
kevinkr said:When you use a grid choke you also end up with a subsonic high q resonance that can be most easily fixed (if you don't like the way it sounds
Hi Kevin. I haven't played with grid chokes yet beyond simulations in LTspice. It revealed the behaviour you described but I was really surprised to see the magnitude of that resonance if the previous plate load was also high impedance, in this case an ideal current source. The peak (from memory, I'm on the road for a couple weeks) was 20+ dB below a couple hertz. The latter depends of course on the sizes of the coupling cap and choke. Juggling common part sizes it would have been easy to place it square in a vinyl/tonearm resonance.
The ways I played at 'fixing' it was series and/or parallel resistance with the choke. Both of course to some degree compromise the benefits of using a choke: series the quick recovery from capacitor blocking and parallel the low load on the driver. On the plus side I have read a post or two here of using parallel resistance and coupling cap values to 'tune' the resonance in such a way to fill in the bottom octave below the roll off of the output transformer. Add in cost, potential for induced hum and finicky tuning requirements, I came away with the impression grid chokes are best suited to follow resistor plate-loaded stages and the primary benefit was low DC grid impedance to ground for quick overload recovery. Does that jibe with your experiences on the bench?
Did you use grid Choke then ???
The value of my orginal Grid res is 320k not 220k as i mentioned.
I put choke and resistor series to ground,so that grid see the choke first ,that caused my whole psu drift??? +/-50 Vdc.
Im using +/-370 - 274b - 6 uf -Choke - 47uf - two 10h hammond chokes parallel- 200uf- 5k primary imp .Then i get about 390 vdc out
Other circuit is standard JeLabs 300b.
Papparazzi
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.