K&K Audio / Lundahl Transformers

Welcome! Need support, you got it. Or share youe ideas and experiences.

Return to K&K Audio / Lundahl Transformers

K&K Audio / Lundahl Transformers

HF rolloff with LL1660S

193.41.233.240

[ Follow Ups ] [ Post Followup ] [ Thread: Display Email Next ] [ K&K Audio / Lundahl Transformers ]

Hello,
I have a DIY KT88 PP amp with per channel two pairs of KT88 in triode.
In the first version I use a LL1660/18ma as IT Splitter, the second version has the new LL1660S/10ma.
The first version with the LL1660/18ma sounds very good,
the new version with the LL1660S/10ma has more bass but also high frequency rolloff which is very audible.
As driver I use both triodes of a 5670 in parallel, 160v on the plates and 7ma through each triode, so 14ma goes through the LL1660.
The configuration of the interstage is different for both versions:
LL1660/18ma in 4:2.25+2.25: this configuration can't be found in the datasheets, it's a combination of Alt-S in http://www.lundahl.se/pdfs/datash/1660s.pdf for the primaries and the secondaries are configured for PP
primaries all in serie (alt-S): Bplus-4, 3-5, 2-7, 8-10, 9-anode
secondaries for PP: negativebias1-13, negativebias2-18, 16-grid1, 21-grid2
Primary inductance approx 70H
source impedance 7.5k
Primary DC current for 0.9 Tesla 16 mA
LL1660S/10ma in 2.25:2+2: this configuration is like Alt-B in the datasheet http://www.lundahl.se/pdfs/datash/1660.pdf
Primary inductance 42H
source impedance 3.5k
Primary DC current for 0.9 Tesla 18 mA
Does anyone know why I have this HF rolloff with the LL1660S ?
And why do I get more bass with the LL1660S even though its inductance is lower than the LL1660?
Regards,
Danny

Follow Ups:

Re: HF rolloff with LL1660S - KevinC 05:04:44 09/12/03 (3)
My speculation is that the essentially 1:2 connection you are using with the LL1660S/10mA means that the paralleled 5670 simply hasn't a low enough driving impedance to prevent high frequency roll-off. My experience has shown me that step-up IT connections only "work sonically" if the driving tube has a very low plate impedance ( <1000 ohms). The increased bass response perception may be just that. With the absence of some high frequency level you perceive more bass.
If you can sacrifice the gain, try the LL1660S/10mA connected as 2.25+2.25:2+2 to see what happens. Use only half of the 5670.

Kevin Carter
K&K Audio
www.kandkaudio.com
- Re: HF rolloff with LL1660S - Danny T 06:15:01 09/12/03 (2)
  Thanks.
  I think I'll revert back to the LL1660/18ma in the 4:2.25+2.25 configuration, this gives me just enough gain.
  Putting the primaries all in serie with the LL1660S 4.5:2+2 will give me not enough gain and in this configuration the LL1660S is rated 10ma so I can use only one half of the 5670 and this will double the plate impedance.
  - Re: HF rolloff with LL1660S - KevinC 06:24:54 09/12/03 (1)
    It does double the plate resistance compared to the LL1660/18mA implementation, but is still effectively 2 times lower than your current LL1660S/10mA implementation, so it is a "compromise" between the two that you may want to listen to just to see.
    
    Kevin Carter
    K&K Audio
    www.kandkaudio.com
    - Re: HF rolloff with LL1660S, solution found - Danny T 13:36:31 11/26/03 (0)
      I've found the reason for the high frequency rolloff:
      see other message:
      http://www.AudioAsylum.com/forums/KandK/messages/1453.html (Open in New Window)

This post is made possible by the generous support of people like you and our sponsors:

Atma-Sphere Music Systems, Inc.

You can not post to an archived thread.