@Tom_S said:
Looking good, Bill! Your power supply is probably the biggest ever used with this little kit. I'm curious to see your measurements. Have you done the feedback and other mods yet? I see the Wima caps.
I have not done the feedback or cathode bypass cap mods yet. I have only replaced the 1uF ceramic coupling caps with 3.3uF WIMA polyesters. This significantly improved the low frequency extension. Other mods that I have done so far involve NOT applying power to the blue LED diode circuit board traces and also NOT applying power to the volume control on/off switch circuit board traces. Only well filtered DC is applied to the tube PCB. I will be putting a separate AC mains on/off switch on the front panel near the transformers, where it can do no harm.
My frequency response measurements are looking good so far. I used my function generator and scope to do spot checks at three frequencies: 20Hz, 1kHz, and 20kHz. I tested the output driving into 3 different loads: 10K, 47K, and 100K. I also tested driving into several different cable capacitances: 84pF, 201pF, 400pF, and 945pF. RCA cable lengths ranged from 3 to 12 feet. What I found is a relatively flat frequency response from 20Hz to 20kHz, regardless of load and cable capacitance. Even the 945pF cable test was flat out to 20kHz driving into a 10K, 47K, or 100K load. The only significant difference is that overall preamp gain dropped from about 14dB when driving into a 47K or 100K load, down to about 11.5dB when driving into a 10K load.
So my conclusion thus far is that the WIMA 3.3uF caps helped to flatten the response down to 20Hz, even driving into a 10k load. And the output impedance, while somewhat high without feedback or bypass, is still capable of driving into almost any normal RCA cable capacitance without significant high frequency loss.
@PWRRYD said:
Yeah the feedback circuit and the cathode bypass caps will make a huge improvement by reducing distortion and reducing the gain.
I agree. Gain is probably a little excessive, ranging from 11.5 to 14dB, depending on load. As soon as I get my Arta jig and REW moved over to my test bench and setup to do distortion measurements, I will run an HD comparison with and without feedback and bypass cap mods. Should be fun. But it will take me a while to get this set up, as I am currently on the very low end of the learning curve using REW with my ARTA jig.
I'm back to tinkering with the preamp kit. Since adding the feedback from the output, I wondered about forgoing the cathode bypass caps. Easy enough to test how much they are helping to lower the THD.
@6thplanet said:
Cathode bypass caps usually boost gain.
Yep, you're right. I had a senior moment there
On a side note, you also could measure the voltage drop across the cathode resistor and then replace it with the proper LED or string of LEDs to get the same drop. I have never done that but have read some posts that claim that LEDs are the hot ticket.
This case has been on my shelf for decades and I think might be perfect for this project. My goal is to use an Arduino Nano board and an IR receiver to drive a motorized volume pot scavenged from an old Sony HT receiver.
@6thplanet said:
Cathode bypass caps usually boost gain.
Yep, you're right. I had a senior moment there
On a side note, you also could measure the voltage drop across the cathode resistor and then replace it with the proper LED or string of LEDs to get the same drop. I have never done that but have read some posts that claim that LEDs are the hot ticket.
That's a good idea. Hmmm....the kit came with 2 blue LEDs that I'm not using.
I know different color LEDs have different forward voltage drops. There is a huge thread over at DIYAudio (search "Red Light District") where the amp designer goes into the benefits of LEDs and how to test different ones for better impedance. That project was obviously a push-pull tube power amp so he had to use a huge array of LEDs to handle the higher current but I've also seen little preamps use one or two LEDs to bias the tubes. Regardless it seems like a fun thing to try with your project and you and Bill seem to enjoy experimenting. Plus LEDs are super cheap and tiny (easy to fit).
No glass in the front on this thing. I'm still trying to decide if I want the tubes right up front. If I do go that route, I'll have to make something like a stepped wall with a "floor" with holes for the tubes.
No idea about the quality of this PE motorized controller, but it is one of the newest of the "New Products" at Parts Express. I have one coming on Monday although I'm not sure yet what I will do with it.
I removed the cathode bypass caps - gain is now down to 5.8dB with the volume pot wide open. That's about perfect. Distortion seems fine. 0.02 - 0.03% at unity gain.
I'm reading one of the shorter LED threads on DIYAudio. Lots of back and forth and questioning others claims. Someone did mention using a voltage reg as CCS. I find that intriguing too.
I think this will work pretty well. I put a little light in the back to simulate the glow, but it’s a little much. Trying to decide if I’m going to paint the insert panel black or polish it.
Thinking about trying the diode bias mod. I measured the cathode voltage drop across my set of four 6J1 tubes. Line voltage was 120vac for all tests. B+ was 58.3vdc for all tests. But because the tube heaters were connected in series, the heater voltage varied with transconductance.
Kit 1, Tube A: 2800 micromhos: 0.379v drop across 200 ohm cathode resistor (1.89ma) Heater: 5.8vdc
Kit 1, Tube B: 3550 micromhos: 0.466v drop across 200 ohm cathode resistor (2.33ma) Heater: 6.15vdc
Kit 2, Tube A: 4500 micromhos: 0.538v drop across 200 ohm cathode resistor (2.69ma) Heater: 6.25vdc
Kit 2, Tube B: 3450 micromhos: 0.488v drop across 200 ohm cathode resistor (2.44ma) Heater: 6.15vdc
It's hard to say, but it looks like I have 1 weak tube, two moderately weak tubes, and 1 good tube. Tom, have you tested your tubes yet to see how they measure? What kind of cathode drop are you getting?
I finished my cabinet today. I drilled a hole in the plastic front panel for the volume control, but, as you can see, the knob only slides about half way onto the knurled shaft. So I will need to re-drill and move the PCB forward about 1/4" to correct this mistake.
I also mounted an on/off switch on the far left side and then cut a rectangular hole on the far right side so that I could see the tubes glowing. Very dim in operation, but it serves as a pilot light. Doesn't look too bad, considering that I only paid $5 for this entire case, transformer, diodes, and first two cap stages!
Sounding good so far, Tom. I need to get a testing fixture setup with my laptop & Steinberg UR22 so I can start doing some measurements. Then I can look at the harmonic distortion percentages of 2nd, 3rd, 4th, etc., when driving into various loads and cable capacitances.
The knob was given to me by @kenrhodes. This is one of the knobs that was on the old 1940's Williamson 807 tube amp that he gave me several years ago. I will be replacing the pots on that amp when I rebuild it with solid shaft types.
Interesting. Looks like my tube tester estimates micromhos a little bit higher than your tester. I used the chart settings for a 6AK5, which may account for the difference.
Comments
Yeah the feedback circuit and the cathode bypass caps will make a huge improvement by reducing distortion and reducing the gain.
I have not done the feedback or cathode bypass cap mods yet. I have only replaced the 1uF ceramic coupling caps with 3.3uF WIMA polyesters. This significantly improved the low frequency extension. Other mods that I have done so far involve NOT applying power to the blue LED diode circuit board traces and also NOT applying power to the volume control on/off switch circuit board traces. Only well filtered DC is applied to the tube PCB. I will be putting a separate AC mains on/off switch on the front panel near the transformers, where it can do no harm.
My frequency response measurements are looking good so far. I used my function generator and scope to do spot checks at three frequencies: 20Hz, 1kHz, and 20kHz. I tested the output driving into 3 different loads: 10K, 47K, and 100K. I also tested driving into several different cable capacitances: 84pF, 201pF, 400pF, and 945pF. RCA cable lengths ranged from 3 to 12 feet. What I found is a relatively flat frequency response from 20Hz to 20kHz, regardless of load and cable capacitance. Even the 945pF cable test was flat out to 20kHz driving into a 10K, 47K, or 100K load. The only significant difference is that overall preamp gain dropped from about 14dB when driving into a 47K or 100K load, down to about 11.5dB when driving into a 10K load.
So my conclusion thus far is that the WIMA 3.3uF caps helped to flatten the response down to 20Hz, even driving into a 10k load. And the output impedance, while somewhat high without feedback or bypass, is still capable of driving into almost any normal RCA cable capacitance without significant high frequency loss.
I agree. Gain is probably a little excessive, ranging from 11.5 to 14dB, depending on load. As soon as I get my Arta jig and REW moved over to my test bench and setup to do distortion measurements, I will run an HD comparison with and without feedback and bypass cap mods. Should be fun. But it will take me a while to get this set up, as I am currently on the very low end of the learning curve using REW with my ARTA jig.
Cathode bypass caps usually boost gain.
I'm back to tinkering with the preamp kit. Since adding the feedback from the output, I wondered about forgoing the cathode bypass caps. Easy enough to test how much they are helping to lower the THD.
Yep, you're right. I had a senior moment there
On a side note, you also could measure the voltage drop across the cathode resistor and then replace it with the proper LED or string of LEDs to get the same drop. I have never done that but have read some posts that claim that LEDs are the hot ticket.
This case has been on my shelf for decades and I think might be perfect for this project. My goal is to use an Arduino Nano board and an IR receiver to drive a motorized volume pot scavenged from an old Sony HT receiver.
That's a good idea. Hmmm....the kit came with 2 blue LEDs that I'm not using.
I know different color LEDs have different forward voltage drops. There is a huge thread over at DIYAudio (search "Red Light District") where the amp designer goes into the benefits of LEDs and how to test different ones for better impedance. That project was obviously a push-pull tube power amp so he had to use a huge array of LEDs to handle the higher current but I've also seen little preamps use one or two LEDs to bias the tubes. Regardless it seems like a fun thing to try with your project and you and Bill seem to enjoy experimenting. Plus LEDs are super cheap and tiny (easy to fit).
Cool! I like your remote volume idea. Is that a glass window on the front panel, so that you can see the tubes glowing during operation?
No glass in the front on this thing. I'm still trying to decide if I want the tubes right up front. If I do go that route, I'll have to make something like a stepped wall with a "floor" with holes for the tubes.
No idea about the quality of this PE motorized controller, but it is one of the newest of the "New Products" at Parts Express. I have one coming on Monday although I'm not sure yet what I will do with it.
https://www.parts-express.com/Line-Input-Passive-Motorized-Preamp-Board-with-Remote-6-12-VDC-320-6012?quantity=1
I removed the cathode bypass caps - gain is now down to 5.8dB with the volume pot wide open. That's about perfect. Distortion seems fine. 0.02 - 0.03% at unity gain.
Awesome!
I just noticed this at PE -
https://parts-express.com/Line-Input-Passive-Motorized-Preamp-Board-with-Remote-6-12-VDC-320-6012?quantity=1
I just posted a link 50 minutes ago (4 posts above).
So you did! How did I miss yours??
Cause I'm an idiot sometimes - that's how.
Neat, I'll have to look into the LED biasing deal.👍🏼
I'm reading one of the shorter LED threads on DIYAudio. Lots of back and forth and questioning others claims. Someone did mention using a voltage reg as CCS. I find that intriguing too.
I think this will work pretty well. I put a little light in the back to simulate the glow, but it’s a little much. Trying to decide if I’m going to paint the insert panel black or polish it.
Thinking about trying the diode bias mod. I measured the cathode voltage drop across my set of four 6J1 tubes. Line voltage was 120vac for all tests. B+ was 58.3vdc for all tests. But because the tube heaters were connected in series, the heater voltage varied with transconductance.
Kit 1, Tube A: 2800 micromhos: 0.379v drop across 200 ohm cathode resistor (1.89ma) Heater: 5.8vdc
Kit 1, Tube B: 3550 micromhos: 0.466v drop across 200 ohm cathode resistor (2.33ma) Heater: 6.15vdc
Kit 2, Tube A: 4500 micromhos: 0.538v drop across 200 ohm cathode resistor (2.69ma) Heater: 6.25vdc
Kit 2, Tube B: 3450 micromhos: 0.488v drop across 200 ohm cathode resistor (2.44ma) Heater: 6.15vdc
It's hard to say, but it looks like I have 1 weak tube, two moderately weak tubes, and 1 good tube. Tom, have you tested your tubes yet to see how they measure? What kind of cathode drop are you getting?
I haven't tested them yet. Too much other stuff to deal with until this afternoon. I'll check the voltage on the original set in a few minutes.
I must have gotten really lucky. I have 0.402v drop on both stock tubes.
I finished my cabinet today. I drilled a hole in the plastic front panel for the volume control, but, as you can see, the knob only slides about half way onto the knurled shaft. So I will need to re-drill and move the PCB forward about 1/4" to correct this mistake.
I also mounted an on/off switch on the far left side and then cut a rectangular hole on the far right side so that I could see the tubes glowing. Very dim in operation, but it serves as a pilot light. Doesn't look too bad, considering that I only paid $5 for this entire case, transformer, diodes, and first two cap stages!
So with only 2 mA of bias current and 0.5ish voltage drop across the cathode resistor I don't think LED bias is even an option. Oh well...
You got me beat on the budget case and parts, Bill! I like that knob too. It reminds me of a B&W TV we has when I was a kid. How does it sound to you?
Yeah, LEDs are out of the running. Hmmm...maybe a Schottky diode with the right voltage drop. Does that sound plausible?
How about this one? See figure 13 in the datasheet.
https://www.mouser.com/ProductDetail/STMicroelectronics/STPS1L60RL?qs=JquAdH79urZpCHmVrbMOgw==
Sounding good so far, Tom. I need to get a testing fixture setup with my laptop & Steinberg UR22 so I can start doing some measurements. Then I can look at the harmonic distortion percentages of 2nd, 3rd, 4th, etc., when driving into various loads and cable capacitances.
The knob was given to me by @kenrhodes. This is one of the knobs that was on the old 1940's Williamson 807 tube amp that he gave me several years ago. I will be replacing the pots on that amp when I rebuild it with solid shaft types.
I tested the 6J1s and 5654Ws that I have on hand.
From what I can gather for my Hickok 532, the minimum for a good tube = 1500 micromhos.
Kit 1, Tube A - 2750
Kit 1, Tube B - 2550
Kit 2, Tube A - 2100
Kit 1, Tube B - 2525
NOS GE Jan5654W, Amazon - 2550
NOS GE Jan5654W, Amazon - 2700
NOS GE Jan5654W, My stash - 1850
Interesting. Looks like my tube tester estimates micromhos a little bit higher than your tester. I used the chart settings for a 6AK5, which may account for the difference.