Little Dot MK8SE / MK6 Super Mods (All verified mods are on first page)
May 9, 2018 at 1:28 PM Post #3,317 of 4,154
IMG_9218.JPG
Correct, sorry if I mislead you when I gave you those resistor readings, they were the measured values, the board values are sometimes different as you say.

Glad to see you are making progress with the repairs, looks like you've got those eyelets in and traces fixed nicely!
I just found out the original WIMA caps install from factory is incorrect and load of garbage component they did put in!
What I have seen from MK8SE layout , all those 15 pieces of red WIMA caps should be model: WKP10 0.22uF/250V, however they put 11 pieces MKP10 0.22uF/250V; 3 pieces MKP10 0.22uF/160V & 1 MKP4 0.1/630V.
I guess may be they out of stock some caps so just find other look similiar garbage put into the PCB. Although the design of this amp is very great but the job they've done and most of component include PCB board out from Little Dot company is rubbish!
Be aware & better double check all those red WIMA caps !!!
 
Last edited:
May 9, 2018 at 1:54 PM Post #3,318 of 4,154
I'm still struggling with schematics, I'm finding it difficult to see what is in parallel to what frankly. I don't see a 390K there, could you elaborate?



I tried measuring from the points in red.
R16 and R17 are feedback resistors. I put in 80k as that’s what the low gain setting is. 390k in parallell with 100k. Doesn’t matter which phase you measure. Input grids are pin 1 and 4. Don’t remember which one is ”negative” and ”positive”, should be easy to trace which grid belongs to which output. Then just measure from a input grid to the output of that same phase.
 
May 10, 2018 at 11:57 PM Post #3,319 of 4,154
Wow alot passed me by I was away.
I will try to help.

This picture with circle red color is the worst issue, that area was where explosion happened, I believe that damage area cause ground shorted
No that is the resistor which burn as a result from either the diode failure or the power transistor.
Regardless, once this happens, it is very likely you will need to replace the associated power transistor that the (zener) diode was regulating.
My stong suggestion is to upgrade BOTH the "PNP" &"NPN" transistors (on the chasis) a set with the upgraded version as those stock units are not made anymore but replaced with higher amperage units.
Also replace both the (100v) zener diodes and the resistors next to them with higher wattage as that I already mentioned is the hottest area of the amp (besides the sockets).


Edit: can you correct a possible misconception, is going from 100K to, say, 50K lower, or higher, I might be using the wrong terminology
Screenshot_2016-04-05-15-01-50.png

The switch is a double pole (two circuit),
Single throw (on off)

If you look carefully, you will realize that switching "on",
will place an added resistor in parallel.
An added paralleled resisitor will drop the total resistance.

Lowering total resistance will increase feedback,
AND then, that
Increased feedback will end up give you a "cleaner" LOWER volume output.
This is "NFB".
NFB is not liked because, although it corrects some distortions, it is a compromise solution for this application (in an amp)
So,
Less feedback equals louder volume.
More feedback will reduce volume.

Remember, more feedback (less volume) is a result of that lower resistance that feeds it..

(I had to do this anyway because I stupidly shorted out the amp when taking voltage readings with a shaky hand
If amp is not functional, You most likely caused the stock "power transistors" to blow.
Shorting "Should" not have damaged the diodes, but I would definitely upgrade them as a damaged transistor can definitely blow the diode..
Just swapping these out remove the necessity of headaches and prolonged testing .
Athough testing is supposed to be a necessary tool to learn the PSU.
I believe this thread had extensive posts on PSU problems, in past.

Higher feedback resistance is lower NFB and lower feedback resistance is higher NFB. Its logical when you think about it :) If there’s lower resistance more of the signal gets fed back, hence higher NFB.
I believe he not fully grasp because
You forgot to mention the volume into how the unit behave for complete picture to understand.
The louder setting is less feedback, so more resistance to achieve that setting.

May I know how many watts recommendation for the resistor to put there?
Quarter watt, just like mentioned and what you removed.

In both cases you'll be more than fine with 0.25W resistors. However I recommend simply using 1W no-name carbon film resistors for all spots that dissipate under 0.75 W. That covers usually 95% of resistors inside any circuit. Why mess with different types
That is an excellent idea,
In general,

But NOT for this board, as the resistors are spaced too close together, and no way are they going to fit.

Not only that, but the leads from the resistors won't fit into the tiny holes...

Plus, the reccomended resisitors to place here are film resistors.
Preferably dale.

I will never ever recommend carbon resisitors because when you look at the data sheets, they have highest noise of all type resisitors.

You will NEVER see carbon resisitors used in volume comtrols.
They will always use best.

Depending on application you may never notice,
But even in high power applications, the best is wirewound.
The question is would it matter, and probably wouldn't in many instances.

In this amp that is confined with heat I would not recommend.
Also remember these old school designs are admittedly very dependant on components selection and quality in certain areas.

In most other amps I would not disagree with your choice, in high watt or PSU areas.


As I am lack of electronic experience especially calculation,further more there are too many choices of resistor to replace and confuse about which is the best. Anyway, thanks for the answer
Use quarter watt high quality resisitors here.
Larger will not fit in the hole.

I'm still struggling with schematics
Screenshot_2016-04-05-15-01-50.png

You may have different values, but same circuit trace.

Be aware & better double check all those red WIMA caps !!!
Unfortunately you pulled all those out, and I not sure if they re-usable,
Buy there "purpose" was not critical, if you implememted the cap mods.

They are parasitic caps to keep circuit clean and "may" have helped with transients as they buffered the PSU.
They are not expensive so its ok.
Its just that you are changing way more than necessary,
And you may not notice much.

Also, I would encourage you to change the (switch area) resistors values a bit...
On the MK8, it is agreed that it sounds better on high gain,
Which is less feedback thru the higher resistor.
I would encourage you to try slightly higher resistance if you like.
You would get more gain as well.

On the MK6, high gain setting is so high, that it is reported to not sound as good, probably from too high resisitor, introducing distortion from to low feedback.

Yet, I am assuming that could be remedied by a larger grid resistor at the powertubes.
Then you could implement less feedback.
Correct me if I am wrong (MrCurwen & Sonic & Coin).
I am guessing that the power tubes may end sound with a bit more harmonics (more euphoric, thicker), but the bias mods may also alter that.

Anyways, maybe I should try myself hopefully will have time this month.
 
Last edited:
May 11, 2018 at 3:06 AM Post #3,320 of 4,154
You will NEVER see carbon resisitors used in volume comtrols.
Hificollective sells stepped attenuators with Takman carbon films. They're 2% though compared to 5% you normally see in carbon films. I use metal film 1% in my stepped attenuator but the rest of my amp (modern LTP) is full of carbon films and don't generate noise. I have some small noise that can only be heard with my sensitive LCD-XC's. Amp is completely quiet with LCD-3 and speakers. It's some kind of ground problem. Will try matching input transformers next, actually have all the parts but I've been busy and amp sounds great as it is so. I'm rambling :)

Yet, I am assuming that could be remedied by a larger grid resistor at the powertubes.
Not really, a larger grid resistor will not lower the gain but potentially hurt transient response if too large. Remember the WCF output stage already has negative gain. I think the stock 300R is about right.

Anyway, great to see you posting again Maxx!
:)
 
Last edited:
May 11, 2018 at 5:08 AM Post #3,321 of 4,154
Wow alot passed me by I was away.
I will try to help.


No that is the resistor which burn as a result from either the diode failure or the power transistor.
Regardless, once this happens, it is very likely you will need to replace the associated power transistor that the (zener) diode was regulating.
My stong suggestion is to upgrade BOTH the "PNP" &"NPN" transistors (on the chasis) a set with the upgraded version as those stock units are not made anymore but replaced with higher amperage units.
Also replace both the (100v) zener diodes and the resistors next to them with higher wattage as that I already mentioned is the hottest area of the amp (besides the sockets).




The switch is a double pole (two circuit),
Single throw (on off)

If you look carefully, you will realize that switching "on",
will place an added resistor in parallel.
An added paralleled resisitor will drop the total resistance.

Lowering total resistance will increase feedback,
AND then, that
Increased feedback will end up give you a "cleaner" LOWER volume output.
This is "NFB".
NFB is not liked because, although it corrects some distortions, it is a compromise solution for this application (in an amp)
So,
Less feedback equals louder volume.
More feedback will reduce volume.

Remember, more feedback (less volume) is a result of that lower resistance that feeds it..


If amp is not functional, You most likely caused the stock "power transistors" to blow.
Shorting "Should" not have damaged the diodes, but I would definitely upgrade them as a damaged transistor can definitely blow the diode..
Just swapping these out remove the necessity of headaches and prolonged testing .
Athough testing is supposed to be a necessary tool to learn the PSU.
I believe this thread had extensive posts on PSU problems, in past.


I believe he not fully grasp because
You forgot to mention the volume into how the unit behave for complete picture to understand.
The louder setting is less feedback, so more resistance to achieve that setting.


Quarter watt, just like mentioned and what you removed.


That is an excellent idea,
In general,

But NOT for this board, as the resistors are spaced too close together, and no way are they going to fit.

Not only that, but the leads from the resistors won't fit into the tiny holes...

Plus, the reccomended resisitors to place here are film resistors.
Preferably dale.

I will never ever recommend carbon resisitors because when you look at the data sheets, they have highest noise of all type resisitors.

You will NEVER see carbon resisitors used in volume comtrols.
They will always use best.

Depending on application you may never notice,
But even in high power applications, the best is wirewound.
The question is would it matter, and probably wouldn't in many instances.

In this amp that is confined with heat I would not recommend.
Also remember these old school designs are admittedly very dependant on components selection and quality in certain areas.

In most other amps I would not disagree with your choice, in high watt or PSU areas.



Use quarter watt high quality resisitors here.
Larger will not fit in the hole.



You may have different values, but same circuit trace.


Unfortunately you pulled all those out, and I not sure if they re-usable,
Buy there "purpose" was not critical, if you implememted the cap mods.

They are parasitic caps to keep circuit clean and "may" have helped with transients as they buffered the PSU.
They are not expensive so its ok.
Its just that you are changing way more than necessary,
And you may not notice much.

Also, I would encourage you to change the (switch area) resistors values a bit...
On the MK8, it is agreed that it sounds better on high gain,
Which is less feedback thru the higher resistor.
I would encourage you to try slightly higher resistance if you like.
You would get more gain as well.

On the MK6, high gain setting is so high, that it is reported to not sound as good, probably from too high resisitor, introducing distortion from to low feedback.

Yet, I am assuming that could be remedied by a larger grid resistor at the powertubes.
Then you could implement less feedback.
Correct me if I am wrong (MrCurwen & Sonic & Coin).
I am guessing that the power tubes may end sound with a bit more harmonics (more euphoric, thicker), but the bias mods may also alter that.

Anyways, maybe I should try myself hopefully will have time this month.
I decide replace all those red WIMA caps with Mundolf caps, but still not sure which model is the best fit?
I see board show 2k, 68k, 100k, 470k near the switch, any recommendation resistor value for that area? I also decide to replace them with Dale
IMG_9220.JPG
IMG_9221.JPG
 
Last edited:
May 11, 2018 at 10:18 AM Post #3,322 of 4,154
the rest of my amp (modern LTP) is full of carbon films and don't generate noise
Yes I agree your amp is an excellent example of design choices to offset need better or expensive parts.

Not really, a larger grid resistor will not lower the gain but potentially hurt transient response if too large. Remember the WCF output stage already has negative gain. I think the stock 300R is about right.

Anyway, great to see you posting again Maxx!
Dam, do you know what this means?.....

I may have to give up my higher gain tubes for the lower "mu" versions...

I was using 6SL7 high gain equivalent in the 6C8G.
The good thing is that there are more of the 6sn7 gain type.
I will swap a 6F8G type there, as I like the bottle shape and they are similar performance.
You liked that high "MU" mullard type,
and I bet you could switch amp to high gain with normal "MU" tube.

There is no reason why your loaded MK6 cannot reach sonic clarity of your new amp.
Although I do not know how "colored" it is, the realism should still be high.
I doubt the mk6 as colored as a 300b amp.
Probably a bit more than a 2a3 amp.

I decide replace all those red WIMA caps with Mundolf caps, but still not sure which is the best fit?
The best fit would be same style WIMA, and high pulse type.
Look in data sheets for best "pulse" capacitor.
They will look exactly the same.
I have seen blue caps in other designs which may be even higher voltages, but I don't know as I haven't researched it.

I see board show 2k, 68k, 100k, 470k near the switch, any recommendation resistor value for that area
For optimal feedback resisitor, that may depend on tube selection.

Yet,
One thing already noted is that the MK8 sounds good on "high gain" selection, which, at 100k, is similar to the MK6 preferred "low gain" select of (80k)..

Let's look at the resulting resistance when switch closed:
Paralleled:
Mk6 390k & 80k = 68k
Mk8 100k & 68k = 40.5k

So the two Settings of the switch Results are:
Mk6 = 390k & 68k
Mk8 = 100k & 40.5k

I Notice my preferred setting of higher gain (less feedback) is very close to Sonic preference of low gain (more feedback) .
They are at similar setting, wish we both prefferred over other setting.

So I would try Having 100k for both,
As the default(combined resisitors) lowest point,
instead of the lower values of 40k & 68k, which result in way higher NFB...

Then you notice 390k is a point debatable as being prefferred in the mk6 as a choice.
That value is up in the air, and may be too high.

So up to you what to use, but what looks to be the same for both amps,
Is to get a combined low of around 100k.
That would be a good start point as the most NFB to use.

Also, for the other (high) setting,
A useful compromise of 300k (not as high on MK6) should be a good starting point to try.

So For both MK6 & MK8,
I would actually use same resisitors to try first.

Try 300k for the higher spot(close to switch), and 150k for the lower spot.

For MK8, the high gain setting may sound higher, but is unknown how clear it would get as these are different tubes.
I feel this is a good start point to try,
Because actuall stock values are set higher in MK6 and debated as too much.

Overall, changing resisitors to try these values will still get you the desired less feedback for both amps.

Any suggestions welcome here.
The future would be to try to eliminate the NFB more,
but that may require more changes in tube selection & tube settings.
So a whole new set of Problems, lol.

I have not even been able to get my CCS installed yet, and I would still have to "dial in" the sweet spot setting I have now.

Hope all that long post helps.
I hope to start working on projects again this month.
 
Last edited:
May 11, 2018 at 10:40 AM Post #3,323 of 4,154
Yes I agree your amp is an excellent example of design choices to offset need better or expensive parts.


Dam, do you know what this means?.....

I may have to give up my higher gain tubes for the lower "mu" versions...

I was using 6SL7 high gain equivalent in the 6C8G.
The good thing is that there are more of the 6sn7 gain type.
I will swap a 6F8G type there, as I like the bottle shape and they are similar performance.
You liked that high "MU" mullard type,
and I bet you could switch amp to high gain with normal "MU" tube.

There is no reason why your loaded MK6 cannot reach sonic clarity of your new amp.
Although I do not know how "colored" it is, the realism should still be high.
I doubt the mk6 as colored as a 300b amp.
Probably a bit more than a 2a3 amp.


The best fit would be same style WIMA, and high pulse type.
Look in data sheets for best "pulse" capacitor.
They will look exactly the same.
I have seen blue caps in other designs which may be even higher voltages, but I don't know as I haven't researched it.


For optimal feedback resisitor, that may depend on tube selection.

Yet,
One thing already noted is that the MK8 sounds good on "high gain" selection, which, at 100k, is similar to the MK6 preferred "low gain" select of (80k)..

Let's look at the resulting resistance when switch closed:
Paralleled:
Mk6 390k & 80k = 68k
Mk8 100k & 68k = 40.5k

So the two Settings of the switch Results are:
Mk6 = 390k & 68k
Mk8 = 100k & 40.5k

I Notice my preferred setting of higher gain (less feedback) is very close to Sonic preference of low gain (more feedback) .
They are at similar setting, wish we both prefferred over other setting.

So I would try Having 100k for both,
As the default(combined resisitors) lowest point,
instead of the lower values of 40k & 68k, which result in way higher NFB...

Then you notice 390k is a point debatable as being prefferred in the mk6 as a choice.
That value is up in the air, and may be too high.

So up to you what to use, but what looks to be the same for both amps,
Is to get a combined low of around 100k.
That would be a good start point as the most NFB to use.

Also, for the other (high) setting,
A useful compromise of 300k (not as high on MK6) should be a good starting point to try.

So For both MK6 & MK8,
I would actually use same resisitors to try first.

Try 300k for the higher spot(close to switch), and 150k for the lower spot.

For MK8, the high gain setting may sound higher, but is unknown how clear it would get as these are different tubes.
I feel this is a good start point to try,
Because actuall stock values are set higher in MK6 and debated as too much.

Overall, changing resisitors to try these values will still get you the desired less feedback for both amps.

Any suggestions welcome here.
The future would be to try to eliminate the NFB more,
but that may require more changes in tube selection & tube settings.
So a whole new set of Problems, lol.

I have not even been able to get my CCS installed yet, and I would still have to "dial in" the sweet spot setting I have now.

Hope all that long post helps.
I hope to start working on projects again this month.


From the picture below, can i replace those red colour circle resistors with 150R/5W instead of 120R/5W & replace those yellow colour circle resistors with 2K2/5W wirewound instead of 3K3/2W?
IMG_9221A.jpg
 
May 11, 2018 at 10:51 AM Post #3,324 of 4,154
Yes it's not critical.
Going up to 150 from 120 means an easier surge current for caps when turn on.


The 3k3 was lowered to 2k2 for less volt drop and more voltage to circuit.
I am unsure if these are the resisitors the same
we talked about earlier in thread ,
That both me and sonic lowered it to 1k (!)
Ask Sonic or Barron as I been away too long.
 
Last edited:
May 11, 2018 at 11:05 AM Post #3,325 of 4,154
Yes it's not critical.
Going up to 150 from 120 means an easier surge current for caps when turn on.


The 3k3 was lowered to 2k2 for less volt drop and more voltage to circuit.
I am unsure if these are the resisitors the same
we talked about earlier in thread ,
That both me and sonic lowered it to 1k (!)
Ask Sonic or Barron as I been away too long.

I have just upgraded those Diode:
Green circle 1N4001 replace with 1N4007
Yellow circle 5.1V replace with 5.1/5W
Blue circle 15V replace with 15V/5W
Red circle 36V replace with 36V/5W
1N4148 next to Yellow circle replace with 1N4007
Those 2 Diodes near transistors 100V/1W replace with 100V/5W


IMG_9221B.jpg
 
Last edited:
May 11, 2018 at 11:51 AM Post #3,326 of 4,154
What I have seen from MK8SE layout , all those 15 pieces of red WIMA caps should be model: WKP10 0.22uF/250V, however they put 11 pieces MKP10 0.22uF/250V; 3 pieces MKP10 0.22uF/160V & 1 MKP4 0.1/630V.
I guess may be they out of stock some caps so just find other look similiar garbage put into the PCB. Although the design of this amp is very great but the job they've done and most of component include PCB board out from Little Dot company is rubbish!

What's the issue here? They're all the same capacitance, and even the same type.

That is an excellent idea,
In general,

But NOT for this board, as the resistors are spaced too close together, and no way are they going to fit.

Oh alright. Well, luckily 0.25W is more than plenty.

PCBs, what a pain. Always go P2P.

I will never ever recommend carbon resisitors because when you look at the data sheets, they have highest noise of all type resisitors.

You are probably thinking about carbon composition, which is the retro type resistor that has high noise and drift. I am talking about carbon film, which is modern type and has no mentionable noise and no drift. Completely different types.

On the MK6, high gain setting is so high, that it is reported to not sound as good, probably from too high resisitor, introducing distortion from to low feedback.

Yet, I am assuming that could be remedied by a larger grid resistor at the powertubes.
Then you could implement less feedback.
Correct me if I am wrong (MrCurwen & Sonic & Coin).
I am guessing that the power tubes may end sound with a bit more harmonics (more euphoric, thicker), but the bias mods may also alter that.

I am asking in all honesty, please explain your reasoning as to why this would be the case. Understanding your preconceptions would help a lot as to how to approach explaining things.

I am especially referring to the role of the power tube grid resistor here.

Dam, do you know what this means?.....

I may have to give up my higher gain tubes for the lower "mu" versions...

Why?

There is no reason why your loaded MK6 cannot reach sonic clarity of your new amp.

There's plenty of reasons it cannot reach the THD, the transient response or the output impedance of that amp. What this means in terms of 'clarity' is up to debate. My own view is that all of these things (low THD without gNFB, first class transient response, very low output impedance) constitute 'clarity'.

I doubt the mk6 as colored as a 300b amp.
Probably a bit more than a 2a3 amp.

SE, balanced, gNFB / open loop...? AC filaments or regulated filaments?

One thing already noted is that the MK8 sounds good on "high gain" selection, which, at 100k, is similar to the MK6 preferred "low gain" select of (80k)..

Knowing one resistor in the NFB loop doesn't mean anything if you don't know the other one. They form a voltage divider. One resistor is in series and one is in parallel, forming a voltage divider. Unless one of these resistors is the same value in both cases, they cannot be directly compared like this.

Overall, changing resisitors to try these values will still get you the desired less feedback for both amps.

I am not sure less gNFB is necessarily desireable in this amp. Lower NFB means higher output impedance, meaning less control over the load. This would bring out the differences of headphones by exaggerating their flaws.
 
May 11, 2018 at 12:00 PM Post #3,327 of 4,154
What's the issue here? They're all the same capacitance, and even the same type.



Oh alright. Well, luckily 0.25W is more than plenty.

PCBs, what a pain. Always go P2P.



You are probably thinking about carbon composition, which is the retro type resistor that has high noise and drift. I am talking about carbon film, which is modern type and has no mentionable noise and no drift. Completely different types.



I am asking in all honesty, please explain your reasoning as to why this would be the case. Understanding your preconceptions would help a lot as to how to approach explaining things.

I am especially referring to the role of the power tube grid resistor here.



Why?



There's plenty of reasons it cannot reach the THD, the transient response or the output impedance of that amp. What this means in terms of 'clarity' is up to debate. My own view is that all of these things (low THD without gNFB, first class transient response, very low output impedance) constitute 'clarity'.



SE, balanced, gNFB / open loop...? AC filaments or regulated filaments?



Knowing one resistor in the NFB loop doesn't mean anything if you don't know the other one. They form a voltage divider. One resistor is in series and one is in parallel, forming a voltage divider. Unless one of these resistors is the same value in both cases, they cannot be directly compared like this.



I am not sure less gNFB is necessarily desireable in this amp. Lower NFB means higher output impedance, meaning less control over the load. This would bring out the differences of headphones by exaggerating their flaws.
Among within those 15 pieces of red WIMA caps, 1 of them is different model (MKP4 instead of MKP10) & another 3 pieces are different power rate 160V (should be at least 250V)
 
May 12, 2018 at 7:14 AM Post #3,328 of 4,154
Try 300k for the higher spot(close to switch), and 150k for the lower spot.
.....
Overall, changing resisitors to try these values will still get you the desired less feedback for both amps.

Any suggestions welcome here.
The future would be to try to eliminate the NFB more,
but that may require more changes in tube selection & tube settings.
So a whole new set of Problems, lol.

Hi Maxx!

My listening testing has been delayed due to PSU upgrades to put right the short I had during measurements, and other problems with the ammeters and gain switch. I've also had a broken leg of one of the Jupiter caps with all the messing around inside so when this phase is finished I don't expect to do anything more....probably!

However I have got one result and that is that the high gain setting sounds very good to me, I did not hear any deficiencies in sound and have therefore decided to keep it on high gain from now on so I don't necessarily agree that the MKVl feedback values mean having to use low gain. I will only change back now if I have tubes or headphones which sound bad with high gain.

My 390K resistor is in fact 330K and I've been looking into removing the switch completely to keep the high gain setting with the help of Sonic and MrCurwen.

I wonder if bloodhawk could try high gain with his tube/headphones and see what he thinks

I have not even been able to get my CCS installed yet, and I would still have to "dial in" the sweet spot setting I have now.

Neither have, I was hoping to have it done by now but I've said why I've been delayed.


From the picture below, can i replace those red colour circle resistors with 150R/5W instead of 120R/5W & replace those yellow colour circle resistors with 2K2/5W wirewound instead of 3K3/2W?

Yes, replace the yellow colour resistors with 2K2, or less, Maxx, Sonic and myself have all gone to around 1K for more voltage and without issues so far.
 
May 12, 2018 at 3:46 PM Post #3,329 of 4,154
What's the issue here? They're all the same capacitance, and even the same type.
of them is different model (MKP4 instead of MKP10) & another 3 pieces are different power rate 160V (should be at least 250V)
This is the difference... The be "MKP" designation, which is the ability for pulse voltages, and that translates for us to have better transients delivery in the PSU.
Even when same spec type (MKP), the higher voltage unit is better.
This is usefull for designs that are reliant on the parts performance.


PCBs, what a pain. Always go P2P.
That is a goal I want to do in a future amp I still hope to do.

I am talking about carbon film, which is modern type and has no mentionable noise and no drift. Completely different types.
Excellent info to know.

I am especially referring to the role of the power tube grid resistor here.
Yes that was a brain-fart in my rusty thinking. The grid resistor was error to think of.

For the usage if the amp in a higher gain setting (of lower NFB), I would complimet with use of a tube with less gain (MU), so wont distort the driver stage.
In other words, I am compensating and adjusting one thing for another.
I know my amp sounds better with less NFB.
I rolled many many 12pin and 6pin in both high and low settings, to get my "fav" tube choice.
It turned outmy FAv tube is an 6C8G in higher gain(less NFB),
and I do not know how much less NFB will sound good, untill I try.


There's plenty of reasons it cannot reach the THD, the transient response or the output impedance of that amp. What this means in terms of 'clarity' is up to debate. My own view is that all of these things (low THD without gNFB, first class transient response, very low output impedance) constitute 'clarity'.
I agree it is not a clear cut simple thing.
For instance, I have heard some tube amps sounding:
"euphoric",
"colored",
"dirty",
"clean",
"lifeless"....
Yet all had high resolve and realism...


SE, balanced, gNFB / open loop...? AC filaments or regulated filaments?
Getting into specifics do not discount the generality of a 300b tube amp still having more harmonics over a 2a3 tube amp ,
because they are all still using old-school traditional designs, unlike your unique modern approach.
That does put a spotlight, on the fact that there are rarely any new design approaches to tube amp design.
This is a sad reality.
Your design is so far the only new or different approach I have seen.

Although the implementation of new parts, like CCS, Gyro, and now switching PSU are new tech,
but they are still being implemented, into what is essentially still all old-school designs(in amps).

they cannot be directly compared like this.
Yes they can as the MK6&8 have same driver tube circiut and I already noted the resistances.
Also I took note already the added paralelled resistor only added when switched in for more fedback.
I figured that the existing amounts were way tooo low a resistance, as the higher gains all sounded better.

I am not sure less gNFB is necessarily desireable in this amp. Lower NFB means higher output impedance, meaning less control over the load. This would bring out the differences of headphones by exaggerating their flaws.
I do agree that this would increase amp output impedance, but unknown for now how much and to what extent.
The WCF output stage basically is in charge of lowering impedance, so how it handles lower NFB is to find out.

so I don't necessarily agree that the MKVl feedback values mean having to use low gain. I will only change back now if I have tubes or headphones which sound bad with high gain.

My 390K resistor is in fact 330K and I've been looking into removing the switch completely to keep the high gain setting with the help of Sonic and MrCurwen.
This is good to know, as that means we can try higher values for the resistors (less NFB).
I would assume that one of the sonic changes of less NFB is a more tube harmonics,or thicker sound, so I would be on the lookout for this effect.
Also the amp output impedance as MrCurwin stated.

Yes, replace the yellow colour resistors with 2K2, or less, Maxx, Sonic and myself have all gone to around 1K for more voltage and without issues so far.
Ok, so have you verified this is the resistor location right?
I would definitely reccommend going to 1k...
 
May 12, 2018 at 4:24 PM Post #3,330 of 4,154
If you look carefully, you will realize that switching "on",
will place an added resistor in parallel.
An added paralleled resisitor will drop the total resistance.

Ah, yes now I see the parallel connection, of course!


On the MK6, high gain setting is so high, that it is reported to not sound as good, probably from too high resisitor, introducing distortion from to low feedback.

I didn't hear any distortion at all on high gain.

Yes I agree your amp is an excellent example of design choices to offset need better or expensive parts.

There is no reason why your loaded MK6 cannot reach sonic clarity of your new amp.
Although I do not know how "colored" it is, the realism should still be high.

This is interesting, the amp manages to sound very clear yet at the same time has a wonderful euphonic type sound, that has been mentioned before on this thread. I don't think this euphonic, or euphoric sound is distortion, or feedback going by what has been said about this previously on this subject so I can't account for how the amp manages to sound so good frankly.
 
Last edited:

Users who are viewing this thread

Back
Top