I would prefer to use take the feedback after the output cap to save one cap and still blocking DC to the grid.Circuit 2:
Last edited:
I successfully converted my 71a to 2p29l. Haven’t listened to it yet since I haven’t had a moment to myself with kids, work etc
But things are looking very good. This tube is very easy to sett up. Way way way less microphonic, which was my main complaint with the 71a. Settings are CCS loaded 22ma (I’m aware a tad high), 170 B+, bias - 5.5V, anode ~130V. Distortion also lower than the 71a.
Also, noting that for me I’m not getting 120ma at 2.2V on the filament as the data sheet shows. Rather 135ma, so had to knock it down with resistor and now have 125ma. This was the case for all 3 tubes I since tested. Not a big deal but threw my filament calculations off- fortunately had some low value decent resistors laying around.
I’m excited to hook it up in the coming days when I have some time to listen. Will update soon!
But things are looking very good. This tube is very easy to sett up. Way way way less microphonic, which was my main complaint with the 71a. Settings are CCS loaded 22ma (I’m aware a tad high), 170 B+, bias - 5.5V, anode ~130V. Distortion also lower than the 71a.
Also, noting that for me I’m not getting 120ma at 2.2V on the filament as the data sheet shows. Rather 135ma, so had to knock it down with resistor and now have 125ma. This was the case for all 3 tubes I since tested. Not a big deal but threw my filament calculations off- fortunately had some low value decent resistors laying around.
I’m excited to hook it up in the coming days when I have some time to listen. Will update soon!
Last edited:
Thanks to these posts by Anchan and euro21 I was finally able to find the source of the low level 60hz hum from my headphone amp. I built a 6sn7-2a3 years ago but didnt like it. I finally settled on a 26-2a3 with coleman filament regs but it had the hum, heard when there is no music playing. Recently I decided to find where the hum was so I improved the power supply, no effect. Removed the power transformers and put them in separate chassis, no effect. Removed the first stage, no hum. so now i know it was coming from the 26 stage, so i replaced the filament regs with a PQ5EV5 series reg and it still hummed. I now suspected the tube so i replaced it but still hums.
From euro21’s post i wrapped aluminum foil on the 26, and finally no hum! I will try the 2P29L, as I dont like the look of the 26 wrapped in foil… ;-)
Haha yes, I myself learned about it from euro21 in another thread. Not sure why no one mentions this. From Ale’s site you would never know it’s even an issue. But I tried eight different 71a’s and all of them had this problem.
Btw, the 2P29L still benefits from tying the metal shield to chassis. Noticeably the tube is much quieter this way. Right now I just have a wire taped to the metal body, but will come up with a better solution. But even without that drain wire this tube is orders of magnitude less noisy and less microphonic. I think it’s also way less ancient than the 26 for instance.
Btw, the 2P29L still benefits from tying the metal shield to chassis. Noticeably the tube is much quieter this way. Right now I just have a wire taped to the metal body, but will come up with a better solution. But even without that drain wire this tube is orders of magnitude less noisy and less microphonic. I think it’s also way less ancient than the 26 for instance.
I collected more types of -mainly- thoriated tungsten filament tubes for decades.
For example in case of #26 I have (few dozens of) tubes from several manufacturers and from several periods.
I measured all of them, and select its according to later use.
In every case I measured them in "live" operating point, and -at least at low input volume- watched the FFT.
The FFT measuring filters out the increased hum, or tendency to microphony (knocking).
Nowadays DIY-ers buy from ebay seller the "pair", and trust that it is appropriate.
The usually tube tester device can't measure such parameters, that bent of radiated hum, or microphony.
If the seller haven't measured it, you can't be sure that this problem may or may not occur.
For example in case of #26 I have (few dozens of) tubes from several manufacturers and from several periods.
I measured all of them, and select its according to later use.
In every case I measured them in "live" operating point, and -at least at low input volume- watched the FFT.
The FFT measuring filters out the increased hum, or tendency to microphony (knocking).
Nowadays DIY-ers buy from ebay seller the "pair", and trust that it is appropriate.
The usually tube tester device can't measure such parameters, that bent of radiated hum, or microphony.
If the seller haven't measured it, you can't be sure that this problem may or may not occur.
I finally got this all wired and am listening to it the last few days. Wow I really love it so far.
I did end up trying a bit of feedback, around 6db. I think it sounds wonderful. I don’t find the life is sucked out or anything like that.
Used 430k through a 1uF into a node with 51k input resistor.
I tried it without as well. I liked it too. I don’t know. Gain structure much better this way though for my system, and to me sounds real nice. Channel matching got better, noise and distortion all went down. YMMV try it out if your curious.
I’m using CCS and had around 15db gain. Now around 9 with the NFB.
I’m coming from a Salas 6V6 preamp, built as per his post 1 in that epic thread. This sounds a bit clearer, and maybe a bit more 3D. Take all these subjective descriptions with a grain of salt. Bottom line though this preamp sounds great. Go ahead and build if you’re on the fence. Microphonics not an issue at all.
I did end up trying a bit of feedback, around 6db. I think it sounds wonderful. I don’t find the life is sucked out or anything like that.
Used 430k through a 1uF into a node with 51k input resistor.
I tried it without as well. I liked it too. I don’t know. Gain structure much better this way though for my system, and to me sounds real nice. Channel matching got better, noise and distortion all went down. YMMV try it out if your curious.
I’m using CCS and had around 15db gain. Now around 9 with the NFB.
I’m coming from a Salas 6V6 preamp, built as per his post 1 in that epic thread. This sounds a bit clearer, and maybe a bit more 3D. Take all these subjective descriptions with a grain of salt. Bottom line though this preamp sounds great. Go ahead and build if you’re on the fence. Microphonics not an issue at all.
@FlaCharlie Could you please share your 1626 preamp schematic? If you also have LTSpice sim file then that would be even better.
What are the effects of running the tube too hot?
I am testing at 175 B+, 18-19mA (10m45s CCS) , ~5V filament bias, about 130V on the anode. g2 and g3 tied to anode with 100R.
Sounds really great, but after about 30-40m running, I get a spike that pops up without fail at 18 kHz at around 90dBV. Not the end of the world, but running it into the power amp it is 20dB higher. I know our sensitivity at that frequency is down a lot, but still.
It always takes about that amount of time, and it's always the same frequency spike. I'll try backing down to 15ma or below. I know Ale running these at 20ma, but maybe it's a tad hot.
It must be a heat issue, if it only pops up after running for a while, right?
I am testing at 175 B+, 18-19mA (10m45s CCS) , ~5V filament bias, about 130V on the anode. g2 and g3 tied to anode with 100R.
Sounds really great, but after about 30-40m running, I get a spike that pops up without fail at 18 kHz at around 90dBV. Not the end of the world, but running it into the power amp it is 20dB higher. I know our sensitivity at that frequency is down a lot, but still.
It always takes about that amount of time, and it's always the same frequency spike. I'll try backing down to 15ma or below. I know Ale running these at 20ma, but maybe it's a tad hot.
It must be a heat issue, if it only pops up after running for a while, right?
Last edited:
I use a simple resistor load with filament bias and I've not had a problem. I've used a maximum of 15mA. I take the black top off - it just unscrews. Don't know if that helps at all. Sounds good in this setup. Not quite as good as a 112A but easier to use and more common. I ground the cover and the pins that should be connected to ground. Attached circuit but with Rod Coleman older V4 regs. The V9 regs need a supply of 14.5V not 17.5V. The output feeds my 2a3 stage hence the 0.1uF/470K.
Last edited:
