|Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum|
Please consider donating to help us continue to serve you.
Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
||Thread Tools||Search this Thread|
|15th April 2010, 03:47 PM||#1|
Join Date: May 2004
Location: Toronto, ON
Taking my KT88 SE E-linear amp to the next level (long)
Not sure what the best choice of topic was for this thread, but I have a number of questions regarding my first tube amp build.
The amp is question is Pete Millett’s Single-ended E-linear Amp.
I built the amp last year and have enjoyed it a lot. One of my audiophile buddies said it was the best tone that he had ever heard in an amp. The bass is a bit lacking though compared to other amps we tried (quite a few). That may just be the SE amp. It also could be that my 90-91db 4-ohm MTM speakers aren’t the easiest/benign load.
I created this thread, as I have some questions for the more experienced builders/designers out there. I’d really like to take the performance up a notch, by first optimizing the performance of the topology as is, then possibly exploring some modifications to the topology (maybe experiment with triode strapping the KT88).
1) First, the biggest issue I have with the amp is a very low level of hum. Pete mentions on his site that he had a bit of hum as well and thought that spacing the choke and the output transformers further apart might have been a better idea. I mimicked Pete’s layout but put a bit more spacing between the iron. I also did not use Pete’s pcb ground board idea and instead used local star grounding but following the current flow that the pcb would force.
When I first built the amp, I had terrible hum until I realized that I forgot to ground the filaments (left floating) as I was contemplating how I wanted to do it and forgot altogether. I grounded it and that cut the hum down substantially but still it was apparent. Then I cut the tie to ground at pin 4 of the D3A and instead created a center tap to ground with two resistors that made a little bit of a difference. Next, I elevated the heaters to ¼ B+ and that dropped it some more. Still, I can hear some hum if I am near the speakers or the room is very quiet.
I need to measure the noise at the output, I know. I will, but with my meter, the reading wasn’t very stable, so I have to fire up the scope.
I have some further idea/questions about how to make it quieter
Is this just the price to pay for AC heating the filaments? Is the D3A with so much gain more sensitive to heater noise than another type of tube? Would AC heating likely make the amp dead quiet?
The other question is about the ultra-path like cap (C1 || C5) on the output. Broskie raises some concern about noise performance of the ultra-path topology (link below). Would bypassing R7 likely quiet the amp considerably by increasing the ripple rejection? Any idea how much gain the KT88 has when in Ultra-linear configuration?
Ultrapath = Ultra Simple...Except that it's not
Lowering SE Amplifier Noise (page 2)
The Tube CAD Journal: E-Mail, UltraPath (right column)
2) Secondly, I would like to make the bass a bit more solid. The bass is not bad, but firming it up would take the amp to the next level. It could be that this is the limit of SE, but I tend to think I can get a bit more.
I did not use the James PTx as it wasn’t easy at the time to source, and instead found a big Chicago Transformers PT, model PCR-300. This PT has similar specs, but a bit less wattage rating. (550-370-75-0-370-550 @ 300mA; 5v@6a, 6.3vct@5a, 6.3ct@1a). I made up for the lower wattage rating--which is still enough for the design in my estimation—by using a separate filament transformer for the D3As and KT88s. The James is rated at 350-0-350 @400mA but it has a bit more headroom.
One idea I had was to use the 550vac windings and loose the first cap (C7). I estimate this should give me close to the same B+ but now is an L input filter. That should better the bass by increasing the regulation, am I right?
I could also try increasing C4 but then I’d need to simulate what is going on as I just significantly changed the filter. Of course, making it an L input filter does that too.
3) I could experiment with triode strapping the KT88. I could also try fixed bias as I have the 75V tap from which to make an easy bias supply. Both of these significantly change the topology though, so I’d like to leave these for after the above experiments.
I am looking to use these experimentations to learn. I am not dissatisfied with the amp, although I think it has some room to improve (maybe just in my implementation).
|1st June 2011, 04:23 PM||#6|
Join Date: Sep 2009
I am a bit crazy but I would connect a 100K resistor from the KT88 plate to the D3A plate. Give it a try and let us know.
(OK, this is actually cheating, it's called Schade feedback and reduces the output impedance - better damping factor with consequent bass improvement)
You could also try bypassing R7 with a high quality cap of around 100uF value.
Last edited by quikie22; 1st June 2011 at 04:25 PM.
|Thread Tools||Search this Thread|
|Thread||Thread Starter||Forum||Replies||Last Post|
|SE KT88 Amp||Cassiel||Tubes / Valves||50||24th March 2011 09:33 PM|
|help with KT88 / mikaels SE amp||soundbadger||Tubes / Valves||3||1st October 2010 12:08 AM|
|Cheap and Chinese KT88 SE amp||andrew_whitham||Tubes / Valves||46||19th May 2007 12:05 PM|
|Help - CFB in a KT88 SE amp||Drewaudio||Tubes / Valves||4||30th January 2007 05:36 AM|
|SkinnyBoy's elcheapo bad sounding, taking way too long to make towers are FINISHED!!!||SkinnyBoy||Multi-Way||35||31st January 2004 01:48 AM|
|New To Site?||Need Help?|