• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Boyuurange A50 300B Review + Mods Series

Jump to Latest
#24
It will be interesting to see.

You should compare cascode-300b and a cascode-driver-300b if the existing transformers allow.

6SN7 could (squeaky) work without elevating the heaters but I agree adding some elevation is a good idea. Also you could add a CCS on the bottom tube.
 
#25
NickKUK,

You said: "cascode-300b and cascode-driver-300b."
Did you mean a cascode 300b with two 300b tubes in cascode?
That would have a very bad damping factor, the effective rp of the top tube in a cascode stage is very high.

And please explain the advantage of adding a CCS (with a bypass cap, or without a bypass cap) at the bottom Cathode of the 6SN7 Cascode stage?
I really want to know what that will do.

A proper working 6SN7 can have the same current using a self bias resistor, as it does with a CCS.

Does it make a difference if the cathode current is +/-10% different from one 6SN7 to another 6SN7, when using a self bias resistor?

Just my questions.
 
Last edited:
#28
If you take it as given that the design is constrained to an octal socket driving a type 300B, times two channels, you still have some interesting options. With enough signal voltage at the amp's input jacks, the 6SN7s could be paralleled, for minimum distortion and best grid drive. But not much input sensitivity. Maybe enough for modern sources, maybe not quite enough.


Or, one socket could be a 6SL7 for both channels and the other socket a 6SN7 for both channels, maybe as direct coupled cathode followers.


The OP has lots of balls in the air, so may not try to run down all the possibilities the rest of us promote, and may want to move on to the next project. I think Nick is proposing to fix the original second stage, although simply removing the direct coupling to it will fix it much more easily, and adhere to the socket constraints.


All good fortune,
Chris
 
#29
Chris Hornbeck said:
The OP has lots of balls in the air, so may not try to run down all the possibilities the rest of us promote, and may want to move on to the next project. I think Nick is proposing to fix the original second stage, although simply removing the direct coupling to it will fix it much more easily, and adhere to the socket constraints.
Click to expand...

I was thinking: CCS -> Cascode -> Driver -> 300B.

There's lots of options. Some minor, some adding and some replacing. Stephe has already said on the video that the transformers aren't going to be replaced and the existing tubes are the only tubes that will be used.
 
#30
Good day Stephe,

I think one option you missed at 3:31 in your "Tube Amp Full Test" presentation is that the two 33uF caps in series could be changed to two 82uF capacitors which would result in 41uF vs 16.5uF. This would help raise your B+ but you might have to increase the 39 ohm series resistor value. The 39 ohm series resistor is used because the 5Z3P J rectifier is only rated for 4uF input capacitance You removed it at 5:51 of the Basic Mods part 1 presentation), so don't remove it unless you change the rectifier to a different one.

At 21:53 I see the blue 250V coupling caps you mention earlier in "Tube Amp Full Test". It is not clear from the video, but they may well be MKT 250VAC capacitors such as the Epcos B81121 series of capacitors. In which case they are Metalized Polyester film caps which are known for harmonic distortion. They would still be marginal since 250VAC would be about 353VDC.

Looking forward to your next presentation.
 
Last edited:
#32
TheGimp said:
Good day Stephe,
The 39 ohm series resistor is used because the 5Z3P J rectifier is only rated for 4uF input capacitance You removed it at 5:51 of the Basic Mods part 1 presentation), so don't remove it unless you change the rectifier to a different one.
Click to expand...

A big part of that whole mod is replacing the 5Z3PA with a 5AR4 to raise the B+. That resistor is no longer needed. I'm leaving those 33uf caps in place to simplify the needed changes, any improvement would be minor.
 
#33
Re: coupling capacitor distortion. It's worthwhile remembering that a coupling capacitor has no voltage across it in normal, in-band, operation and thus cannot distort the signal. Also, that distortion components 100dB down from a (very loud) 85dB SPL signal are more than 10dB below the noise of our own blood pumping in our bodies, and more than 30dB below any very quiet home.


Filtering caps are potentially significant; coupling caps easily fall into the fashion category.


All good fortune,
Chris
 
#34
Chris, the coupling cap between the driver stage and grid of the 300B has 190V bias across it.

The MKS2 in the article referenced by NickKUK showed -128dB 2nd and -126dV 3rd with no bias, -96dB 2nd and -126 3rd with 18V dc bias, and -92dB wnd and -126dB 3rd with 30V bias.

It is reasonable to expect the distortion to be much higher at 190V bias.

Granted, it is mostly if not all 2nd harmonic distortion and may be masked by the distortion of the 300B, unless it is out of phase in which case it could contribute to produce 3rd harmonic distortion.

Without sufficient measurement in the actual circuit, it is unknown.

I was surprised what I could hear in an ABX test of a high quality audio op-amp with distortion well below -110dB, compared to the same signal bypassing the op-amp and properly gain balalnced. But to hear subtleties I had to really concentrate on identifying the differences through a set of headphones. It totally took the pleasure out of listening to the music.
 
#35
Monolythic op-amps typically have class AB outputs, so aren't inherently monotonic for distortion vs. signal level. I'm not convinced that they're a relevant analogy to the case of coupling capacitors, which can be assumed to be monotonic down to a very granular level. AFAIK.


But the argument remains that if no signal voltage appears across the capacitor, it cannot distort that voltage. Coupling caps, in their working range, don't have signal across themselves. Filtering caps, including RIAA etc. do, and we do need to be concerned about them. Coupling caps are largely fashion bling IMO.



All good fortune,
Chris
 
#36
Good day Chris,

I was trying to point out that it is possible to hear 'some' distortion below -110dB. In the ABX test I hit above 90% confidence which really surprised me. My expectation was that I would not hear anything.

Non-monotonic distortion is certainly easier to hear due to masking of monotonically decreasing distortion.

Ceramic caps (X7R) I have heard distorting. But they are an extreme. Once you go to PIO, metal film, etc I am not sure if anyone can really hear a difference. I may try to set up a test some time for my own edification. Difficult to do unless it is properly matched ABX.
 
#37
Chris Hornbeck said:
Monolythic op-amps typically have class AB outputs, so aren't inherently monotonic for distortion vs. signal level. I'm not convinced that they're a relevant analogy to the case of coupling capacitors, which can be assumed to be monotonic down to a very granular level. AFAIK.


But the argument remains that if no signal voltage appears across the capacitor, it cannot distort that voltage. Coupling caps, in their working range, don't have signal across themselves. Filtering caps, including RIAA etc. do, and we do need to be concerned about them. Coupling caps are largely fashion bling IMO.

All good fortune,
Chris
Click to expand...

I'm playing catchup here - and have a few basic questions/thoughts.

1. Signal across - I'm assuming here that there's no electrical signal across but given the coupling cap is operating via electron attraction/repulsion?
Essentially making an interface - once susceptible to changes in the signal created on the opposite side. Interfaces being notorious for issues.

2. Capacitor structure - given a wound style with edge connection but a single serial wound electrode presents a different frequency impedance vs a stacked plate where each plate is essentially a fork thus current split into separate capacitors. I can see that there would be differences measurable and logically makes sense. This is before we get to the metalised film vs foil etc.
How much plays a difference is both objective on the build (given the cap doesn't sit by itself) and use but also surjective in the sound and enjoyment.
If you have a bad design, it may sound different/better with a $1000 cap vs a $3 cap but in reality you're loosing out due to the bad design.

3. Manufacture tolerances - this is one that I find easy. Given caps like WIMA have to satisfy a specific small % tolerance as sold with a specific defined characteristics vs. premium caps that offer either no guarantee match between components in a batch or 20% for example. It's then up to the designer of the amp to design in the tolerance other wise you're testing every component in every batch with a wider distribution in 'audiophile' vs industrial caps. Not many DIYers will spend $10,000 on finding matching $500 caps vs $20 to find closer matching $3 caps..

In the end just do what sounds better to you.. you don't have to prove anything. "The easiest person to fool is yourself" (Feynman).
 
Last edited:
#38
A couple of thoughts on this in relation to the DUT:

The coupling caps I removed from the amp are clearly forged copies of an ERO cap and who knows what sort of QC they have. I also plan to replace all the cathode bypass caps for the same reason, they likely are forged copies.

They aren't even the capacitance listed in the schematic and on the board (0.56uf instead of the specified 0.47uf), which IMHO were too high a value already. 0.33uf is the max I would want to see in the circuit.

I did make an error assuming the 250V printed on the cap was DC volts. After looking up the specs on a real MKP-1840 the DC voltage rating is over 600V, so it electrically would be OK to leave these in place after jacking up the B+ with a 5AR4 and jumping that 39ohm resistor. That might help some users who don't want to dive that deep into modding the amp.

I'll admit I have a personal preference for the Mundorf Aluminum Oil coupling caps. I've listened to these vs the lower cost solen etc MKP caps and just feel they sound "richer" to me. I don't feel bad splurging $15 for a coupling cap designed for audio, I'm not sure I'm interested in spending more than this though.

I used these lower end Mundorf MKP 0.33uf caps for now, because I have a bunch of them laying around from a previous project. I'm also interested to see how changing these MKPs for the Aluminum Oil caps later will sound.

Then again there might be a placebo effect here? But given the HUGE piles on money I see people dropping on audio stuff, $30 isn't going to get me in a tizzy 😛 People spend 10-20X this much on a few "dark field" cable elevators lol.
 
#39
#40
And here is the final video as a wrap-up with listening impressions and recommendations. There is a point where you have to say "Stop here" and I think these fairly inexpensive mods were a good place to end. Spending over 1/2 of what the Amp costs just doesn't make sense and that's what replacing the output transformers would be, plus I'm honestly not sure there is even room to do that on this chassis. It produces over 2X the power it did as delivered and at 3W the distortion dropped from 3% to well under 1%. Sonically it's much improved and with some easy to drive speakers, is a joy to listen to now.

Boyuurange / Reisong A50 300B Tube Amp Mods: Wrap-up Impressions - YouTube
 
You must log in or register to reply here.
Top Bottom