Cloning the Klein+Hummel UE100

hi gents , I've Got a problem with "baustain A" , it is the first stagage after the imput Amp.
It works, but wen I switch it on the signal gain from -3 dbU to +5dbU, so 8db between input/output are wrong. 
My layout is correct ( checK it a lots of time ) values  are all ok, My conclusion is that the schamatics in the manual wrong or at least incomplete;
CA7, for example, It is in serie with the coupling cap of the previous stage, it cause a roll off around 170Hz, the same is for CA9, so to macke it working I've to remove those both.

the feedback arrangment between, V1 Katodhe PA2 followed by RA11 seams to me not right, trimming  PA2 when switch on 12db slope , there's no differance. what the metter here ?

before to take apart My original unit to check it out. I ned to know if here there's someone that has
experience with this beast. or at least is able to do a correct analysis


http://img535.imageshack.us/img535/2866/baustaina.jpg
Uploaded with ImageShack.us

One thing that doesn't look right to me is RA8; this 1k resistor loads the output of V2, which attenuates it by ca. 50 dB...
According to my schemo, it should be 100k.

thax abbay,
there are 3 AR8 in this schema, ( it is a mistake of mine ) the one you mean are 1M, and when you select one freq. this resistor will be parallel to the resistor of the selected freq. that how the filter works, If I have understed it right.

the problem here is that the wole sistem gain 8db even if I let the switch to 0 position.
so my guest is that the second half of V1 add too mutch gain, tomorrow I'll try to change
the RA8 / RA2 values ( the kathode and plate resistors of the second Half of V1 ) to lineup the sistem to 0db.

apart to this without CA9, and CA7 the sistem works.

Mik.

mik said:

thax abbay,
there are 3 AR8 in this schema, ( it is a mistake of mine ) the one you mean are 1M, and when you select one freq. this resistor will be parallel to the resistor of the selected freq. that how the filter works, If I have understed it right.

Click to expand...

The resistor is the one after the second valve at the output of the 47n capacitor. It is 100k on my schemo, could be 1meg no problem though. Look at the original schemo, this is R89.

the problem here is that the wole sistem gain 8db even if I let the switch to 0 position. so my guest is that the second half of V1 add too mutch gain, tomorrow I'll try to change
the RA8 / RA2 values ( the kathode and plate resistors of the second Half of V1 ) to lineup the sistem to 0db.

Click to expand...

The actual gain of the unit is set by the feedback network, the open-loop gain of the amplifier being almost irrelevant. With the wrong resistor here, the output level should be attenuated by about 50dB. There must be something else that's not correct. Are you sure the output signal is connected to the end of the ladder (the 1.5k resistor at the right end of the gain switch)?

Abby, really thank you for the help me.
you're right the correct value is 100K as you told, it was a transcription mistake. 

about the ladder of the feedback : yes, in the original schema, the string resistors is fron KA2 to KA3, right ?
wen the switch is in "0" db position, KA2 KA3 are shorted eachother , in this condition the original unit have 0 db gain. Mine has +8db completely flat over the whole band.

If you turn right the switch and none of the button are pusched down, the unit boost at 63Hz, exactly what the original unit does, apart from the fact that mine boost from +8db to about +20db instead of 0, +20

moreover, Baustain"B" have tha same amplifier, so at the end of the 2 stages I'v got +16bdU..

I really do not understend..
P.S.
I'm using JJ ECC83s maby does this metter ?

mik said:

Abby, really thank you for the help me.
you're right the correct value is 100K as you told, it was a transcription mistake. 

wen the switch is in "0" db position, KA2 KA3 are shorted eachother , in this condition the original unit have 0 db gain. Mine has +8db completely flat over the whole band.

Click to expand...

This is extremely strange...Can you measure the level at plate of V2, then post 47nF cap? Are you sure there is zero ohm from junction CA2-CA8 and CA6 when the switch is at its end position?

I'm using JJ ECC83s maby does this metter ?

Click to expand...

No; As I said earlier, the system is feedback-controlled hence the whole performance is almost totally independant of the valve parameters.

I told you, this is strange.

The level at the plate is +7.60dbU, after the cap is +8db

resistence between CA2-CA8 is "0"

I have shorted CA6 and nothing has changed... damn!

but for what I have understend this feedback control the amount of boost you can give at a particulary freq.

I'm going to chek the layuot again, who knows..

hi mik,

do you have some infos about the transformers and chokes used in the ue-100. ratios? inductance?

i want to do a clone too, already started to build the in/out stage. but didn't have time to finish it

cheers

christian

form My clone I'm using Sowter 3575 for input, 1120 for output, and 1130 for plate choke.

M.

ok, now all baustain works good except for C and D, they needs some extra work.

a big discovery was the totaly useless of the input amp: it adds alot  more noise and distortion and it lower the headroom, the maximun signal allowed without it,  is around +18dbU, and with it in the path, is just +8du because of his gain. so, to compansate the incoming signal we have to use a pad before the input trafo.. Not the champ !

erervy baustain is a buffer itself, and works great at line level, so why use it ?

about the output amp, I'm plan to use somethig with a better performance and little more modern design.

M.