Since I'm going to create a PCB directly from schematics the PSU section looks a bit odd (with separate RC filters for each tube and with two reservoir capacitors of smaller size because that's what I have avaliable).
The idea is to run the pentodes at slightly over 200V Va, 30 mA Ia.
I plan to omit the NFB from secondary and test it first instead and if I like the sound of it I'm going to leave it that way. I left that part of the circuit inthere just in case (other authors use NFB).
Now the questions:
1: Should I consider putting in a switch that woudl toggle between pentode and triode mode ? I will be using rather inefficient fullrange speakers (86 dB/W/m).
2: Should I consider putting in grid stoppers (and if so, where and what value) ?
3a: Heater supply is AC. Should I reference it to the ground (or some other potential, say Vh/2 + a little above the cathode to avoid diode action between cathode and the heater) using a resistor divider or can it be safely left floating ?
3b: Should I change heater supply to DC instead ?
4: Are there any ungodly stupid mistakes in my schematics that I have missed despite its simplicity ?
5: Any other tips for a relative newbie to tubes ?
Thanks in advance for your help guys
The idea is to run the pentodes at slightly over 200V Va, 30 mA Ia.
I plan to omit the NFB from secondary and test it first instead and if I like the sound of it I'm going to leave it that way. I left that part of the circuit inthere just in case (other authors use NFB).
Now the questions:
1: Should I consider putting in a switch that woudl toggle between pentode and triode mode ? I will be using rather inefficient fullrange speakers (86 dB/W/m).
2: Should I consider putting in grid stoppers (and if so, where and what value) ?
3a: Heater supply is AC. Should I reference it to the ground (or some other potential, say Vh/2 + a little above the cathode to avoid diode action between cathode and the heater) using a resistor divider or can it be safely left floating ?
3b: Should I change heater supply to DC instead ?
4: Are there any ungodly stupid mistakes in my schematics that I have missed despite its simplicity ?
5: Any other tips for a relative newbie to tubes ?
Thanks in advance for your help guys
An externally hosted image should be here but it was not working when we last tested it.
1) The availability pf pentode mode might be in order for 86dB/w speakers, along with NFB of some sort for that matter. Damping and distortion in pentode will likely be unacceptable without cathode and/or global negative feedback.
2) A grid stopper on the input triode is definitely in order to stop RF input if not parasitic oscillation. ~1K. 100+ on the pentode g1 (3 on schematic) might be a good idea, too.
3) Whether DC or AC, ideally one would reference it +30VDCish over ground with a high resistance potential divider, bypassed to ground.
While not strictly necessary in a 2 stage amp, more decoupling between the stages might be nice. You already have caps there (C2), add several K of R (to B+), perhaps increase the caps size.
5) Mind the grounding scheme.
2) A grid stopper on the input triode is definitely in order to stop RF input if not parasitic oscillation. ~1K. 100+ on the pentode g1 (3 on schematic) might be a good idea, too.
3) Whether DC or AC, ideally one would reference it +30VDCish over ground with a high resistance potential divider, bypassed to ground.
While not strictly necessary in a 2 stage amp, more decoupling between the stages might be nice. You already have caps there (C2), add several K of R (to B+), perhaps increase the caps size.
5) Mind the grounding scheme.
Tweeker said:
Thanks for bringing this up - I just realzied this is my first blunder. I intended to lower B+ for triode using a 10-15K resuistor as datasheet suggests only 100V anode voltage for it
So I have to fix that and add 1K grid stopper to triode and 100R to pentode, as you have suggested.
Thanks for your reply, any advice is greatly appreciated !
Tweeker said:
You are right, at Vg = -1V Ia = 0.434 mA and 220K * 0.434 mA = a tad over 95V down from estimated 210V B+. Add a 15K resistor and it's just about there. I copied this bit from some schematics I found online that supposedly function well.
Now that you mentioend this I checked the datasheet again and it made me think. It says operating point of triode section for class A amplifier is Va = 100V, Vg = 0V (?), Ia = 3.5 mA, although U/I curves don't exactly confirm that. Does this mean that I could couple the input directly to triode, omitting C1 ?
If I went with Vg = -0.5V at Va = 100V, that would still put me at safe Ia = 2.1 mA. This would reduce R3 + R4 to 238R and R6 + the resistor I forgot about before above it to ~52K. Any thoughts on this - do I need more power from the first stage ?
Its better to increase R6 rather than increase the decoupling resistor beyond 15K.
The drive from the first stage should be fine, this job is what it was born to do.
My datasheet gives max anode for the triode section as 250V, 100V is just the typical operating value.
The datasheet is saying that the triode @ 100V conducts 3.5ma when the grid is driven to 0. This is not where you'd normally idle it.
Stay with at least 2K on R3.
The drive from the first stage should be fine, this job is what it was born to do.
My datasheet gives max anode for the triode section as 250V, 100V is just the typical operating value.
The datasheet is saying that the triode @ 100V conducts 3.5ma when the grid is driven to 0. This is not where you'd normally idle it.
Stay with at least 2K on R3.
kmaier: current component value choices seem to be in line with schematics other people posted online, plus topology is the same simple two stage amplifier with global NFB, nothing new here
Tweeker: I changed the schematics as per your suggestions. I also changed it into more sensible mono block. What I had before was the same thing with rectifier stuck together for no good reason. I have few more questions:
1: What would be a reasonable value for resistor divider on heater supply (R14 + R15) ? 1K + 1K ?
2: I would prefer to avoid too much current going through R16 + R17, can I go any higher with resistance there ?
3: Should I use higher value grid stoppers and if so, where to put the -3 dB (cut-off) frequency point of RC formed by grid stopper and Cg (just slightly above 20 KHz or considerably further up) ?
4: Should I be concerned with HF decoupling of supply (i.e. adding lower capacity, lower ESR capacitors in parallel to existing electrolytics) ?
Others are welcome to chime in with your experience
Tweeker: I changed the schematics as per your suggestions. I also changed it into more sensible mono block. What I had before was the same thing with rectifier stuck together for no good reason. I have few more questions:
1: What would be a reasonable value for resistor divider on heater supply (R14 + R15) ? 1K + 1K ?
2: I would prefer to avoid too much current going through R16 + R17, can I go any higher with resistance there ?
3: Should I use higher value grid stoppers and if so, where to put the -3 dB (cut-off) frequency point of RC formed by grid stopper and Cg (just slightly above 20 KHz or considerably further up) ?
4: Should I be concerned with HF decoupling of supply (i.e. adding lower capacity, lower ESR capacitors in parallel to existing electrolytics) ?
Others are welcome to chime in with your experience
An externally hosted image should be here but it was not working when we last tested it.
The original psu was just fine, mostly dual mono anyhow, saves money if this matters.
1) A few hundred a piece, not big deal. Bypassing the heater directly to ground with small cap each side (a few nF) if they are handy is good.
2) Triple them is fine.
3) Much further up, you do not want it effecting the pass band at all. At least 100kc.
4) Its not the elcos you want to bypass if desired, its the load. If your bypassing the power section youd mount from location 6 to ground, not across the cap. If you have fancy low ESR elcos the first things to replace are C3 and C4.
5) as mentioned breadboarding is a good idea
1) A few hundred a piece, not big deal. Bypassing the heater directly to ground with small cap each side (a few nF) if they are handy is good.
2) Triple them is fine.
3) Much further up, you do not want it effecting the pass band at all. At least 100kc.
4) Its not the elcos you want to bypass if desired, its the load. If your bypassing the power section youd mount from location 6 to ground, not across the cap. If you have fancy low ESR elcos the first things to replace are C3 and C4.
5) as mentioned breadboarding is a good idea
Follow-up on our previous discussion:
I put together one channel. These are my observations and alterations (schematic depicting actual board is enclosed below):
1: Everything works and I haven't blown anything up yet Attached to computer single channel is plenty loud for my environment even with this inefficient full-range speaker (86 dB/W/m).
2: One of my PCL82s appears to be dead Heater lights up gradually but there is no audio output. I haven't checked whether it is triode or pentode section that is dead ... this is on my todo list.
3: 50/100 Hz hum is next to non-existent (it can only be heard during heating up at point blank range directly in the axis of the speaker and it dissapears once tube starts conducting). I'm really glad I haven't screwed up the PCB design.
4: There is rather audible hiss present that sounds exactly like radio or TV that isn't on a channel. Those of you who have been involved with radio communications (ham, CB, etc.) know what I'm talking about - this is the stuff you usually squelch out, the "white noise" that is always out there. I'm not sure whether it is induced in circuit or sourced by computer as I haven't tried shorting out the input yet (I have everything connected on my desk, unshielded).
5: There is no global NFB connection. Everything is in place but I tried the circuit before I conencted the relevant cables and I thought the sound was decent enough. It is however bass-shy and I'd like to improve that as well.
6: I ended up with 100R grid stoppers (unlike what Tweeker had suggested) 'cause that's what I've found in some other schematics using this particular type of tube. I figured I'd go up in value if I noticed RF interference yet I'm not sure wther my hiss problem is RF-related or not.
7: I was thinking about removing R4 (100R under the triode cathode resistor) to have cathode completely bypassed by C3. This would increase gain slightly, correct ? Would this be sensible, given the low value of R4 compared to R3 ?
8: PSU filter resistor is 220R. Sorry about misleading value in schematic !
9: Odd looking heater grounding is a remnant from previous plan to put heaters at some 35V above ground. I haven't soldered in the other resistor in voltage divider and the accompanying bypass capacitor as I don't appear to have hum problem in the first place. But some paranoia never hurt anybody, right ?
Any tips regarding #4 and #5 ? I would like to have more bass/less treble and I'd prefer that hiss to go away as much as possible.
Schematic:
I put together one channel. These are my observations and alterations (schematic depicting actual board is enclosed below):
1: Everything works and I haven't blown anything up yet
Attached to computer single channel is plenty loud for my environment even with this inefficient full-range speaker (86 dB/W/m).2: One of my PCL82s appears to be dead
Heater lights up gradually but there is no audio output. I haven't checked whether it is triode or pentode section that is dead ... this is on my todo list.3: 50/100 Hz hum is next to non-existent (it can only be heard during heating up at point blank range directly in the axis of the speaker and it dissapears once tube starts conducting). I'm really glad I haven't screwed up the PCB design.
4: There is rather audible hiss present that sounds exactly like radio or TV that isn't on a channel. Those of you who have been involved with radio communications (ham, CB, etc.) know what I'm talking about - this is the stuff you usually squelch out, the "white noise" that is always out there. I'm not sure whether it is induced in circuit or sourced by computer as I haven't tried shorting out the input yet (I have everything connected on my desk, unshielded).
5: There is no global NFB connection. Everything is in place but I tried the circuit before I conencted the relevant cables and I thought the sound was decent enough. It is however bass-shy and I'd like to improve that as well.
6: I ended up with 100R grid stoppers (unlike what Tweeker had suggested) 'cause that's what I've found in some other schematics using this particular type of tube. I figured I'd go up in value if I noticed RF interference yet I'm not sure wther my hiss problem is RF-related or not.
7: I was thinking about removing R4 (100R under the triode cathode resistor) to have cathode completely bypassed by C3. This would increase gain slightly, correct ? Would this be sensible, given the low value of R4 compared to R3 ?
8: PSU filter resistor is 220R. Sorry about misleading value in schematic !
9: Odd looking heater grounding is a remnant from previous plan to put heaters at some 35V above ground. I haven't soldered in the other resistor in voltage divider and the accompanying bypass capacitor as I don't appear to have hum problem in the first place. But some paranoia never hurt anybody, right ?
Any tips regarding #4 and #5 ? I would like to have more bass/less treble and I'd prefer that hiss to go away as much as possible.
Schematic:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.