Aegis DIY Tube Headphone Amplifier
Jan 20, 2024 at 10:41 AM Post #1,711 of 2,386
You can also buy wick that's impregnated with flux, which helps as well.
Yes, yes. That's all I ever use. As well, buy a good brand. Trying to save a few bucks with cheap stuff will result in high decibel utterance of 4-letter words. :smile:

I've had very good experience with the Chemtronics brand. There are others that are good as well, it's just what's worked for me so I always go back to it when I need more.

https://www.amazon.com/gp/product/B00P1WZ70K/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1
 
Jan 20, 2024 at 11:16 AM Post #1,712 of 2,386
Yes, yes. That's all I ever use. As well, buy a good brand. Trying to save a few bucks with cheap stuff will result in high decibel utterance of 4-letter words. :smile:

I've had very good experience with the Chemtronics brand. There are others that are good as well, it's just what's worked for me so I always go back to it when I need more.

https://www.amazon.com/gp/product/B00P1WZ70K/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1

Yeah, Chemtronics and MG Chemicals are my go to. With those, I don't need to add flux. A bit of solder on the tip still helps, though.
 
Jan 20, 2024 at 12:31 PM Post #1,714 of 2,386
Yeah, Chemtronics and MG Chemicals are my go to. With those, I don't need to add flux. A bit of solder on the tip still helps, though.
Totally agree. And one other tip: the good desoldering wicks are braided from pure copper, and will oxidize over time. Best to store unused portions in a ziploc bag (with as much air squeezed out as possible) when not in use which will at least slow down the process.
 
Jan 20, 2024 at 7:31 PM Post #1,715 of 2,386
Had what was a great listening session on the Aegis (with my new DCA E3 cans, future post coming on the synergy), until things went a bit south for me.

First problem sucks, but isn't shocking, second problem surprised me and really sucks.

1) Like @mattiastomas I have discovered a nasty noise issue in my chain when attempting to do some gaming on my PC. No noise at all to insufferable noise the second a game is loaded up; my Liberty DAC II is hooked up via USB from my office PC, so the culprit there is pretty obvious. A bummer, because never had line noise issues with other amps during gaming but I suppose the Aegis is such a beast it's picking it up!

2) The real problem; I have discovered a channel imbalance, on Hi-Z output only. No imbalance on Lo-Z, but Hi-Z has a 5db imbalance in favor of the left channel. I've barely used Hi-Z until today (just to test it with DCA E3), although previously I didn't notice an issue with my ADX5000 on the couple of tracks I tested Hi-Z with.

I'm guessing the problem must be at the impedance switch, but any feedback/assistance that can be offered will be gladly taken.
 
Jan 20, 2024 at 7:40 PM Post #1,716 of 2,386
Had what was a great listening session on the Aegis (with my new DCA E3 cans, future post coming on the synergy), until things went a bit south for me.

First problem sucks, but isn't shocking, second problem surprised me and really sucks.

1) Like @mattiastomas I have discovered a nasty noise issue in my chain when attempting to do some gaming on my PC. No noise at all to insufferable noise the second a game is loaded up; my Liberty DAC II is hooked up via USB from my office PC, so the culprit there is pretty obvious. A bummer, because never had line noise issues with other amps during gaming but I suppose the Aegis is such a beast it's picking it up!

2) The real problem; I have discovered a channel imbalance, on Hi-Z output only. No imbalance on Lo-Z, but Hi-Z has a 5db imbalance in favor of the left channel. I've barely used Hi-Z until today (just to test it with DCA E3), although previously I didn't notice an issue with my ADX5000 on the couple of tracks I tested Hi-Z with.

I'm guessing the problem must be at the impedance switch, but any feedback/assistance that can be offered will be gladly taken.

If the imbalance is there on low Z but not on high Z, then it's probably a bad connection on the switch PCB, either a bad solder joint or potentially a short between two pins. The noise is probably through the USB ground, switching to an optical connection may help.
 
Jan 20, 2024 at 7:46 PM Post #1,717 of 2,386
If the imbalance is there on low Z but not on high Z, then it's probably a bad connection on the switch PCB, either a bad solder joint or potentially a short between two pins. The noise is probably through the USB ground, switching to an optical connection may help.

Will try optical on it, or just move the DAC to a standalone streamer entirely. That should fix the issue (I barely play games anymore anyway).

So the channel imbalance is actually only on Hi-Z, no issue on Lo-Z. Still likely the switch PCB?
 
Jan 20, 2024 at 7:50 PM Post #1,718 of 2,386
Will try optical on it, or just move the DAC to a standalone streamer entirely. That should fix the issue (I barely play games anymore anyway).

So the channel imbalance is actually only on Hi-Z, no issue on Lo-Z. Still likely the switch PCB?

Yes, most likely the switch. If it was on both settings, might be a tube, but if only on one there is probably a connection problem on the PCB.
 
Jan 21, 2024 at 4:55 AM Post #1,719 of 2,386
Unbelievable... I swapped the heater supply and the noise switched to the left...
 

Attachments

  • IMG_20240113_175530_heaters.jpg
    IMG_20240113_175530_heaters.jpg
    121.5 KB · Views: 0
Jan 21, 2024 at 6:08 AM Post #1,720 of 2,386
Unbelievable... I swapped the heater supply and the noise switched to the left...
It’s like being an investigator 😀

In the past I used to be a SaaS programmer, risk analytics… coding like this type of problem is a process of careful and meticulous elimination. Eventually, it always leads to the problem. It sounds like you’re getting close now!! 👍🏻
 
Jan 21, 2024 at 6:27 AM Post #1,721 of 2,386
Unbelievable... I swapped the heater supply and the noise switched to the left...

Have the heaters always been wired that close to the rectifier socket? I would keep them away from the rectifier and shorten them. What this sounds like is rectifier noise being coupled into the heater circuit. In Aegis, the heaters in the output stage are referenced to the cathode so the heater-to-cathode voltage will not be violated (although this would only potentially happen near maximum power output). So it's possible this rectifier noise is leaking into the cathode.

That could happen if the wiring is very close to the rectifier, so try moving the wiring away and shortening it if possible. If that doesn't work, it's also possible the coupling is occurring inside the mains transformer itself. If that's the case, another thing you could try: whichever heater winding is contributing to the noise, use that one for the 6SL7s and use the one currently on the 6SL7s for the output stage. The 6SL7 heaters are referenced to ground, not cathode.
 
Last edited:
Jan 21, 2024 at 7:17 AM Post #1,722 of 2,386
Found a LTSpice model of the KT170. At least in a Spice simulation, it biases around 70mA, so right at the upper limit. Using a higher Vdrop rectifier like a 5R4 would take that down to the 60-70mA range. So yes, KT170 will work as well.
@L0rdGwyn just following up on the above...how would the KT170 work with the 274b rectifier? Cheers
 
Jan 21, 2024 at 7:36 AM Post #1,723 of 2,386
@L0rdGwyn just following up on the above...how would the KT170 work with the 274b rectifier? Cheers

Yes it should be fine, 274B is similar to a 5R4. May bias a little high with a lower Vdrop rectifier like GZ34. You can measure the bias current by checking the voltage drop across R2 and R3 and applying Ohm's law.
 
Jan 21, 2024 at 7:49 AM Post #1,724 of 2,386
Yes it should be fine, 274B is similar to a 5R4. May bias a little high with a lower Vdrop rectifier like GZ34. You can measure the bias current by checking the voltage drop across R2 and R3 and applying Ohm's law.

Excellent thanks. I'd like to try some "New Production" KT tubes to see how they compare with the GEC KT88 and I think it'll be between the KT150 & KT170.
 
Jan 21, 2024 at 11:46 AM Post #1,725 of 2,386
It’s like being an investigator 😀

Good that I did this before starting the transplant of components, eh eh

In the past I used to be a SaaS programmer, risk analytics… coding like this type of problem is a process of careful and meticulous elimination. Eventually, it always leads to the problem. It sounds like you’re getting close now!! 👍🏻

I'm a C developer, I do something similar in software. Thanks, fingers crossed!
 

Users who are viewing this thread

Back
Top