Because literally every piece of silicon sold on Aliexpress is fake. That $1 AD797 opamp is an LM741. The 4 amp bridge rectifier has four 1N4003 diodes inside.
Commit heresy and use a series blocking cap. Given the large value feedback resistance required the cap can be pretty small to achieve a sub 1 Hz corner frequency. While to my ear it sounded great the disadvantage is by definition the topology can't correct for problems on the secondary. In my case living with less than maximised 10 kHz square wave and a 20 kHz 1 dB droop was a worthwhile trade off.
In other words while easy to implement success is situational and likely depends on the quality of the OPT.
WHICH high quality parts?
The $500 silver foil caps which short on their own? 🙄
Where users instead of junking them try to rewind them to *maintain their investment?"
Oh well 🙄
Who mentioned AliExpress?
AliExpress is for suckers, karma will hit you hard if you buy a $20 parted by an unknown 8000 miles away for 80 cents. You get what you pay for.
But all the very high quality caps sold by Mouser, Digikey, Farnell, etc. am
or used in commercial good quality Audio of all kinds, smartphones, computers, TVs, etc.are Chinese.
Fwiw their "silicon" too.
You get what you pay for [2]
They use at least Nichicons, Panasonics or Elnas instead of standard electros. There must be a reason behind that, no? Maybe they have more impact on speakers or guitar amplifiers. I'll let you know when my biased ears listen to the difference 😝. Or not.
At least this order looks nice, don't you think?
and this was added later for bypassing either the driver or output stage cathode bypass caps:
And yes, I bought 2 different sets of driver stage cathode bypass caps to see which one I like better. I am one of those on whose good marketing (and some reviews) impacts a lot 😂. Wish me luck!
and this was added later for bypassing either the driver or output stage cathode bypass caps:
And yes, I bought 2 different sets of driver stage cathode bypass caps to see which one I like better. I am one of those on whose good marketing (and some reviews) impacts a lot 😂. Wish me luck!
@grovergardner you were absolutely right. I received the 5AR4 yesterday and that actually was THE gamechanger for now. Now I have all instruments right where they need to be. I don't even know how to describe it. Thanks for that!
@Chris Hornbeck did you see this? would you still recommend changing the feedback resistor from driver's plate to the cathode? I tried that before having the 5AR4 in hand and got me more "power" so to speak, but I found it a little muddy and lacking some detail.Distortion messures (performed by Stephe) where really good as it is now, lower than 2% at 4W:
I know its a lot of fun messing around with a cheap tube amp and gains a lot of experience doing so. If you have money then a better more watts tube amp kit may be a long term aim. I would stay clear of expensive non electrolytic caps as they offer little improvement metal films will do you. If you need a big cap motor run caps are worth considering.
You would apply a resistor from the 8 ohm tap to the cathode of the 12AX7. But unless you absolutely know what you're doing, and have a 'scope to test the amount of feedback and the resulting square waves, I would not recommend doing this. And the chances are very good that slapping in some global feedback will make the amp sound worse, not better.
Well, I did some mods, and then some more, and then more. I ended up with a very satisfying combo:
- used Russian FT-3 for coupling, bypassed by a Duelund JDM 0.1uf
I must say that there is audible difference between caps, it's not snake oil at all. The vast improvement came, apart from the coupling cap, changing the driver stage bypass cap for Audio Note Kaisei. It gave me plenty, but I mean plenty, of more detail.
That's all for now. I'll post if I make further changes. Thanks everyone!!
- used Russian FT-3 for coupling, bypassed by a Duelund JDM 0.1uf
I must say that there is audible difference between caps, it's not snake oil at all. The vast improvement came, apart from the coupling cap, changing the driver stage bypass cap for Audio Note Kaisei. It gave me plenty, but I mean plenty, of more detail.
That's all for now. I'll post if I make further changes. Thanks everyone!!
I have an explanation and developed a theory on capacitor, cable and the audibility of many things considered as snake oil, buy I need more free time and finances to shape it into objective measurements. It has nothing to do with the transfer of electrical signals, but acoustic properties of materials, which reflect on the acoustic output on the speakers.
Hint: Oil acts as dampener
Hint: Oil acts as dampener
In a car suspension or cannon recoil absorber?
Sure enough.
But .... in a capacitor?
No physical movement anywhere, what would it be dampening?
On an Electrical level: wave "dampening" sounds like very lossy dielectric, dampening oscillations.
Not sure that is good.
You've got this exactly right from my own findings with caps. I always use FT-2 and FT-3 for coupling caps. Never tried bypassing them, though. I use a few in parallel for higher values. For cathode bypass I generally use Kemet DC Link caps, but nearly as good was Kaisei 100uF parallelled with Silmic 100uF. Kaisei are about the tops for electrolytics. Somehow putting a Silmic with it was a nice combo. Silmics do need to break in - didn't notice that with the Kaiseis.
1. Always use either 2 Mono Blocks, or 1 Stereo amplifier to do coupling capacitor testing.
(and do the same for electrolytic capacitor testing; stereo amplifiers can only do this with separate bypass capacitors, separate B+ capacitors;
bypass caps that are connect to both channels, and B+ caps that run both channels can not be tested this way).
Put one coupling cap type in one channel; and a different type coupling cap in the other channel
(but with the same capacitance and same voltage rating).
Put 2 identical speakers side by side, one speaker per channel.
Find a real good recording, good on either Left or Right channel. Pick the best channel (L or R).
Feed 1 and only 1 channel output (Left for example), and apply it to Both amplifier channels.
To facilitate easy A B testing, use a switch to feed one channel at a time (instead of having to change balance or individual volume pots).
Listen.
Change recordings
Listen.
Then tell yourself, and us, which capacitor is better.
2. Do Not wire both channels with one type of capacitor pair and listen; then shut down and put in the other type of capacitor pair and listen.
Think about it! That kind of test with human audio memory being what it is, is Not fair.
Do not publish the results of a test that is run that way . . . I will thank you for observing that rule.
Or, if you want to, run all your listening tests using the method in number 2. But be sure to tell us that you used # 2 test method.
Thanks!
(and do the same for electrolytic capacitor testing; stereo amplifiers can only do this with separate bypass capacitors, separate B+ capacitors;
bypass caps that are connect to both channels, and B+ caps that run both channels can not be tested this way).
Put one coupling cap type in one channel; and a different type coupling cap in the other channel
(but with the same capacitance and same voltage rating).
Put 2 identical speakers side by side, one speaker per channel.
Find a real good recording, good on either Left or Right channel. Pick the best channel (L or R).
Feed 1 and only 1 channel output (Left for example), and apply it to Both amplifier channels.
To facilitate easy A B testing, use a switch to feed one channel at a time (instead of having to change balance or individual volume pots).
Listen.
Change recordings
Listen.
Then tell yourself, and us, which capacitor is better.
2. Do Not wire both channels with one type of capacitor pair and listen; then shut down and put in the other type of capacitor pair and listen.
Think about it! That kind of test with human audio memory being what it is, is Not fair.
Do not publish the results of a test that is run that way . . . I will thank you for observing that rule.
Or, if you want to, run all your listening tests using the method in number 2. But be sure to tell us that you used # 2 test method.
Thanks!
Jeez, hadn't realized that I could get to the 470uf for the output cathode bypass using some 220uf Silmics/Cerafines/Muses KZ that I have laying around somewhere. I'll try that if I can get to those 470uf. Thanks for the tip!
I currently have a 470 Mundorf Mcap paralleled with a 0.1uf Jantzen Superior + 1uf Epcos but not sure I like that combo a lot.