Mr Liang 845 Sound :(

[swamp]

Hello all...

Does anyone else have a Mr Liang 845 amplifier and have the same problem as myself and my mate Dave. When we first got these amplifiers they were highly musical with bouncy bass underpining the music.... We did some tube rolling and clarity improved, still maintaining that musicallity.

However about 100 hours use has now passed and both Mr Liang 845 's have gone quite thin and non musical sounding ? This is with 845 original and 845B's

We have tried Lampizator's new circuit which uses 6sl7 / 6sn7 tubes as a replacement driver stage and that bring back the original musicallity. Lukasz says this is because Liang uses Pentode's instead of Triode's.

However I do not want to give up on Mr Liang's original tubes (6ac7 and 5881) and design if it can be brought back from the 'brink'...

Anyone any input / advise.

Regards

David

Circuit here...

http://www.triode-systems.com/uploads/photos/120.png

[Original Burnedfingers]

Could you include voltages at the tubes? That might be a good indicator that something was not correct.

[swamp]

I will have to re-measure them. I cannot remember many off the top of my head. I'll measure them over the weekend.

David

[swamp]

I have to say it did not sound as bad last night. It was a little fuller sounding although still not as good as when new. Normally when tube and caps run in they usually sound better ?

I measured the voltages and they are as shown.

BTW, first tube is a 6ac7, 2nd stage is a 6L6

I am thinking of trying some Valve Art 6550A in place of the 6L6, the power tx can handle the heater demand at 2.5A per side.

Regards

David

[David Robert]

Hi All
I also have a Mr Liang 845, and it seems to be behaving in a similar way to yours Swamp.I have changed coupling caps to obligatto same as you , replaced input stage tubes and changed output tubes to 845b.I thought the amp as recieved from Mr liang was one of the best capable of driving a rythm and making you want to get up and dance along with the music,but to me it always sounded a touch dirty or not quite right,thats the reason I made the mentioned changes.I must admit that sometimes late at night it sounded quite good, but next day back to not quite right.I saw the mods carried out by lampizator guy,and having built one of his tube output stages for my cd94,I thought I would give it a try,the results are a far better than I expected rightness to the sound.I cant explain why the sound is so much better butLampizator says its because he uses triodes and Mr Liang uses pentodes, but I cant help wondering if Mr Liang has not used pentodes as best he could,his circuit looks abit long winded to me.Hope this helps .
David Robert

[swamp]

Since no one has replied with any ideas may be someone can tell me if these will work...

1) take the signal off the anode of the 2nd stage (6l6) rather than the cathode. I have never been too keen on cathode followers. Zout will be higher though, will this work ? Simple case of move the 0.47 cap ?

2) and/or... remove stage 1 and the volume pot and feed signal in on pin 5 of the 6l6 making it a pure power amp.

Regards

David

[kevinkr]

Originally posted by swamp
Since no one has replied with any ideas may be someone can tell me if these will work...

1) take the signal off the anode of the 2nd stage (6l6) rather than the cathode. I have never been too keen on cathode followers. Zout will be higher though, will this work ? Simple case of move the 0.47 cap ?

2) and/or... remove stage 1 and the volume pot and feed signal in on pin 5 of the 6l6 making it a pure power amp.

Regards

David

You can remove the volume pot, but you still need the first stage as it provides all of the voltage gain prior to the output tube. Reconfiguring to get rid of the cathode follower is more problematic.

One of the big problems with some versions of this design is that the cathode follower runs at really low currents. This design seems to run at about 4mA, frankly it seems to me that this is too low for good linearity with the 6L6. I would try modifying to use a 6J5 or similar at about 5 - 6mA which would probably sound a lot better. (Just make sure you don't exceed the maximum voltage ratings of the 6J5 in the redesign.)

You might also take a look at the quality of the coupling caps and electrolytics in the signal path. Replacing a few of these with better parts might help.

FWIW I would be surprised if the 6550 wasn't even worse, this circuit just doesn't run the CF at a reasonable quiescent current for any power tube.. I would consider the KT66 a slightly better choice, but not a good one. (see above)

Overall I wouldn't call this a very good design, but the chassis is nice, and the iron apparently acceptable so it should make a good vehicle for mods. I don't see any reason to keep the original design if you have already identified a mod that sounds significantly better.

In terms of the sound quality I suspect you have gotten used to the good aspects of its performance and are now noticing the warts.. Happens to all of us regardless of the merits of the device in question.

[swamp]

I was considering driving it with a pre-amp if that first stage was removed.

If R106 is lowered to 20 - 25K then that would increase the current on that cathode follower.

Does anyone have any links to recommended 845 schematics worth building ?

Thanks

David

[kevinkr]

Originally posted by swamp
I was considering driving it with a pre-amp if that first stage was removed.

If R106 is lowered to 20 - 25K then that would increase the current on that cathode follower.

Does anyone have any links to recommended 845 schematics worth building ?

Thanks

David

Hi David,

You would need a pre-amp capable of swinging more than 240Vpp cleanly to drive the 845 to full power so you really need that first driver stage.

Yes reducing R106 will increase the current in that cathode follower, but might also result in fried power transformer winding depending on the rating of that particular winding. I would think no less than 20 - 30mA would be required to get the 6L6 up high enough on the transconductance curve to justify its use, and this is clearly a bad idea if you don't know the current rating of that winding.

You could use a 6J5 in the first stage and a 6SN7 configured as an srpp running at about 8mA in the second stage and this ought to work pretty well.

Google for links to 845 amplifier designs, there are plenty out there.

[swamp]

Originally posted by kevinkr


Hi David,

You would need a pre-amp capable of swinging more than 240Vpp cleanly to drive the 845 to full power so you really need that first driver stage.

Yes reducing R106 will increase the current in that cathode follower, but might also result in fried power transformer winding depending on the rating of that particular winding. I would think no less than 20 - 30mA would be required to get the 6L6 up high enough on the transconductance curve to justify its use, and this is clearly a bad idea if you don't know the current rating of that winding.

You could use a 6J5 in the first stage and a 6SN7 configured as an srpp running at about 8mA in the second stage and this ought to work pretty well.

Google for links to 845 amplifier designs, there are plenty out there.

Hi Kevin...

I dropped the cathode R to 24K WW and it sounds much better with more weight. I am sure the tx can cope since it has unused heater taps for feeding a 300B as a driver. The same HT used to drive the 300B is being used to drive the 6l6 in this case, Mr laing uses the same transformers.

Current is now 5.5mA, still not up there but better than 3mA as previous !!

Thanks for the pointer.

Regards

David

David