Matching the KLPP1 pre amp to K5881 MK2

[Richard]

Hi, yes we have the circuit but there was usually a description with these design articles. Have a read through see what it says and if R17 or feedback is mentioned at all. You say the line stage is 3x which looks low so I'd expect something more about it in there.

Save the scan(s) as a jpg or combine them into a pdf if your scanner software does that, then attach them to an email- description now arrived!

[barnesdavid20]

Hi Richard, just been over to Matthews to get a couple of 5751s to try,
Anyway all it says in the article with ref to that area is, I quote:
For the line stage another 6922 section is used. Local feedback is used to reduce gain to approx 3, allowing low output sources to be used, but without having to much gain for CD with its 2v output. it does not actually refer to R14 and R17.
Hope this helps
David

[Richard]

Thanks David that's what we needed!

So line stage gain is lowered by feedback and the only place for that is the node of R17/R14. That divider is interesting in itself as when the vol control is fully down (min vol) the R14 leg will be 180K and when it's fully up it will be 180K + 250K = 430K. So as vol is raised there will be more feedback which will lower gain and thus volume so the control law will be altered I think.

Anyway best not to meddle there The easiest place to attenuate it overall will be, as Andy says, 470K resistors before the pot (if Maplin metalfilm check they give you the right value). That would take care of the line stage gain but I'd stick with the divider on the power amp rather than taking it off and trying and get even more attenuation in the pre as both units will then have usual output/input levels, ie 52dB gain on MM, Zero gain on line, and 300mV sens on power amp.

Another consideration would be to change the line stage to a cathode follower which will have no gain and good drive and may be easier than you think to do. Is it pcb or tag strip construction?

[barnesdavid20]

Thanks Richard, appreciate your help. So I put the resistors in between the pot and the selector switch, will give it a try tomorrow. If that works can i go one step further and change the pot to shunt mode, looking at my pot it dosn't look that difficult, but I assume that i would have to reverse the connections from where they are now, ie the wire going to the selector switch would go to earth and the one going to earth would go to the selector switch.
Have fitted the two 5751s they are jan philips, in place of the 7025s and they definitely sound better, the sound seems a lot cleaner and clearer, also the millivolts across the speakers are identical at 1.2mv, slight hum is there but you have to have your ear touching the speaker grill, so I would think that would be acceptable for an amp with a sensitivity of 75mv. Even though the 5751s are suppose to have 30% less gain, you hardly notice the difference in practise. Really pleased because i seem to be gradually sorting out all my issues, thanks to you guys.
David
PS KLPP1 is built on a printed circuit board, so reluctant to implement to many changes there, was hard enough to fit the large soniqs in the signal path, managed it by leaving the wires to the orange drops in place and soldering to them.

[Richard]

Yes the shunt pot mod will work with the same effect on the feedback circuit, and very similar volume law, to a resistor before the standard pot so give it a go if that gives a suitable level.

All 3 pot connections need changing study the diagram in this thread http://www.world-designs.co.uk/forum...ead.php?t=6885

It may take a bit of getting your head around the drawing but the wiper, 2, moves towards 3 for max volume in both cases.

Note that in the SPM drawing the wiper 2 is to ground so shorts the bottom (as drawn) of the pot track and it only uses the track above the wiper line. I'll try and add that info to the drawing.

[bikerhifinut]

Just read all that.........

Me heid's spinnin'!!

Anyway, Richard seems to have nailed it. When I saw the circuit I thought "Mullard mixer preamp circuit" from the Mullard book. With reference immediately to the line stage, that's how mullard configured the gain stage with series feedback (did I get that right), fed into a CF using the second half of an ECC83, rather than 6922/ECC88.

I'd just stick the 470k in series with the input to the volume pot and see how the attenuation works before any other tweaks. If it sounds ok to you then you've cracked it.

But Richards suggestion of modding the final gain stage to CF really appeals to me. You may have to scratch your bonce a bit but I think it could be done on the PCB as a CF uses very few parts and most are already there. Instantly you have the preamp providing the right amount of gain, i.e unity on line sources and whatever the phono stage makes.
But thats a relatively major fiddle and first off just try a series resistor at the input end of the pot.
Theres also another advantage to Cathode followers and thats the high input resistance/low output resistance which gives good matching to amp inputs and can help to reduce noise/hum etc.
Good to see you are finding a solution though.

A.

[bikerhifinut]

[quote=barnesdavid20;81117]
Have fitted the two 5751s they are jan philips, in place of the 7025s and they definitely sound better, the sound seems a lot cleaner and clearer, also the millivolts across the speakers are identical at 1.2mv, slight hum is there but you have to have your ear touching the speaker grill, so I would think that would be acceptable for an amp with a sensitivity of 75mv. Even though the 5751s are suppose to have 30% less gain, you hardly notice the difference in practise. Really pleased because i seem to be gradually sorting out all my issues, thanks to you guys.
David

I think there won't be any difference in overall gain, the negative feedback values need to be altered to achieve this. Merely swopping a lower mu valve in doesnt guarantee lower gain.
You may also consider trying some different ECC83 as they are similar valves. But it sounds like the 5751 have done the trick for you.

I tried 5751 in a leak Stereo20 and very quickly went back to ECC83's but I never altered the feedback network. I found it confusing, but as I understood it the feedback tries to restore the overall gain to the designed value, but like i said i found it confusing and may have it all wrong.

[barnesdavid20]

Hi Richard, reporting back, have just put the two 470k resistors on the input to the pot and left the amp as is, as you suggested, seems perfect, you now have to turn the volume control just over half way to achieve quite a loud volume, where as before it was only one quarter of a turn. Is this about normal?
If anything the sound seems better, is there any reason for this?
This might seem a silly question but is the line amp doing anything different to what it was before, as it would appear on the face of it that all you are doing is turning the volume up a bit further to get the same input to the line stage valve as before, and as such it is still a x3 amp but you are just able to use more of the pot track to control the usable volume, given this situation does it in any way improve or effect the sound.
Have attached the resistors to the two output pins of the multi input switch so is now easy for me to just change the leads on the pot to do the shunt mod.
As i have now several resistors in the signal path both in the pre and power amp is it worth changing them to a better quality to improve sound and if so what do you recommend?

David

[Richard]

Hi David,

Quite a few points there.
Yes midway position is good as you have expanded the usable range and can now turn it up or down with more control in the evening etc.
The channel balance might be a little better too as before all the attenuation was being done on the tracks at the end of their travel where any error might be worse.

The interesting one though is that you are now running with more feedback and less gain in the line stage for any given loudness. See my post before last about the fb divider leg being connected to the vol pot and therefore feedback increasing and gain decreasing as the pot wiper is raised.
This I would say is a good thing as the designer will have expected the control to be around middle position and folk with MC carts would be running it there. Instead you have been running at the lower end with less fb and more gain. You may now have less distortion and noise at any vol setting than before as the pot is in a different position for the same sound level.

For resistors directly in the signal path I use metalfilms and for high values these may be better with less noise. I have compared them in the shunt pot mod and found a definite difference between carbon film and metalfilm with an Alps blue, the mf sounding better and the combination working very well compared with the pot alone. Others may find differnt but I find an Alps Blue quite muddy on it's own and used to use a Panasonic. As with anything though we can tip things too far so be careful to keep a good voice rather than getting carried away with more detail which will eventually be un-natural and tiring. I tend to use metalfilms in the signal path and carbon films for ground connections though the differnces should only be noticeable in high values and high current positions.

Now for something different, a cathode follower has no gain but is a great buffer providing a high impedance input and low impedance output for good drive. KLPP1 was designed with gain on the line stage to bring phono up to MC level which you don't need. If you find it all works nicely now level-wise have a think about this,

[barnesdavid20]

Hi Richard, thanks for that. Quoting from Morgan Jones: ( The best line amp is no line amp ) he says if you can connect your signal direct to the power amp you will achieve the best possible quality, so are we saying that by taking the 470k out and connecting the output of the pot directly to the power amp, bypassing the line amp altogether that i would achieve an even better sound.
It would also work for phono because it has already been amplified to around 2v by the time it gets to the pot. getting even more confusing now!!!

David