• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Kegger/blueglow kt88 project

Hi everyone, I’ve been building and tinkering with this amp for almost a year now and my tweaking has settled enough that it’s time to share. This is my first scratch build so go easy on me! It’s mostly how Mark built his, but with a few tweaks & design choices.

- Used a 6BZ7 driver tube only because I had one.
- 4 ohm O/T secondaries since my speakers are Magnepan .7’s. Yes I drive Maggies with this amp in a smallish room and it’s magnificent. More on this below.
- Obsessed over the grounding scheme, probably went overboard, but this is a hobby right? The amp is silent. hum is < 1 mVrms.
- On my mains voltage of ~123 VAC the KT88 and 6BZ7 filaments were running at almost 7 Vrms, so I added parallel 0.18 ohm resistors on both sides of the center tap. (Total 4 resistors) for a total of 0.18 ohms series resistance. Now the filaments are bang on 6.3VAC.
- No UL
- No Schade feedback. Using my prized new (to me) HP3580A I confirmed that the Schade feedback decreases 2nd harmonic significantly and higher order harmonics to negligible levels. But to my ears there was no comparison, I guess I like the harmonics. At 1.5W output, 1000 hz, I get ~1% 2nd harmonic & unmeasurable higher order. At 6W, 2nd is 2.5%, 3rd is 0.3%. I left in the feedback wiring unhooked so I’ll probably tweak it to see if I like some higher resistance as a compromise.
- Played around with several KT88 cathode resistors and settled on 560 ohm. DC measurements below.
- Over a few months of use, the fuse blew a couple of times, so I installed a CL-80 thermistor. This fixed it.
- Coupling caps reduced to 0.047MFD. This brings the 3db high pass point up to 15 hz. I run subs with this amp crossed over at ~60 hz so I wanted to remove super low frequency work from this amp. I measured that these O/T’s can’t cleanly pass frequencies below 20-30 hz anyway. This coupling cap change made a huge improvement to sonic quality not only in the low end but way up into the mid-range.
- Tried various reductions on the plate load resistor of the 6BZ7 down to 28k (up to 1.7W plate dissipation). In frequency and distortion sweeps I couldn’t measure an appreciable difference, but in listening tests (subs off) I much preferred the 39k. More holographic all around, less strident midrange. Again, I guess I like the harmonics!
- The KT88s are a bit too close to the O/T transformers. In use the heat seems fine, but I’ll know better for next time!

DC voltages (running on 123VAC mains drawing 1.3A):

- B+ 484, B++ (before the 6BZ7 plate load resistor) 452
- KT88 (L / R): plate 479 / 479, screen 478 / 478, cathode 43.7/44.0. Plate dissipation (W) 33.9 / 34.1
- 6BZ7 (L / R): plate 184 / 175, cathode 2.8 / 2.8, 39k plate load resistor top vs. bottom 268 / 278. Plate dissipation (W) 1.2 / 1.2

I know some folks will be skeptical of running maggies with this amp. I’ve done months of listening comparisons between it and a big Marantz 2285B I restored. Both amps cleanly drive the maggies to loud levels (>85dB in my 18’x15’x9’room). Of course the subs help a lot. But this amp is special. On excellent recordings with a lot of ambient space captured, you’re transported to the room. It’s three-dimensional. You can feel the bow on the string, see the singer’s lips. Wow.

Thank you everyone who contributed designs, videos, and expertise to this thread. It's been super rewarding and fun for me to build!

Colin
 

Attachments

  • 50p_PXL_20240629_223924681.MP.jpg
    50p_PXL_20240629_223924681.MP.jpg
    148.9 KB · Views: 92
  • 50p_PXL_20240629_224209664.MP.jpg
    50p_PXL_20240629_224209664.MP.jpg
    122.7 KB · Views: 92
  • 50p_PXL_20240630_180033209.MP.jpg
    50p_PXL_20240630_180033209.MP.jpg
    146.4 KB · Views: 94
pistachiofil said:
Hi everyone, I’ve been building and tinkering with this amp for almost a year now and my tweaking has settled enough that it’s time to share. This is my first scratch build so go easy on me! It’s mostly how Mark built his, but with a few tweaks & design choices.
- No UL
- No Schade feedback.
Click to expand...
In triode mode, I agree no need for the feedback, in UL you need it.
 
Antonios X said:
I wonder if i can replace the KT-88s with KT-120s without replacing anything else?
Click to expand...
The problem is the increased heater current and if your power transformer has enough headroom. Also you would still be using the same watts of dissipation so the amp wouldn't have any more power without more changes.
 
StepheK said:
The problem is the increased heater current and if your power transformer has enough headroom. Also you would still be using the same watts of dissipation so the amp wouldn't have any more power without more changes.
Click to expand...
I believe I have the necessary power requirements. If I need more power what I have to change in the original schematic?
 
Antonios X said:
I believe I have the necessary power requirements. If I need more power what I have to change in the original schematic?
Click to expand...
If your post from a little ways back is accurate and you have the Hammond 373BX, its 6.3v winding is rated for 5A.
Considering a tungsol KT120 could pull 1.7-1.95 and two would be double that,
Id say you had 1A of headroom.
Id defer to the more knowledgeable folks here, but 1A of headroom might be OK.
I personally might "try it" as 80% of its rated load doesn't sound horrible...
As others have eluded to, with out being able to change the output tubes bias point easily, at most all you will get is a tube whose distortion profile may or maynot sound pleasing to you.
 
Its not really a circuit change other than if I did swap out all the components to handle the extra current, I would want to play with the biasing points to make sure everything is swinging well in the middle of the curves. Its mostly making sure all the components in it can handle the extra current draw. Admittedly, I've not gone down that road and selected parts to accomplish that.
 
Ok. So, you don't think upping the B+ by 5 to 10% with corresponding changes in parts to handle the extra current is necessary to really take advantage of the kt120 higher output capability?
 
I did a pair of KT120 mono amps based around this Kegger design and they sound fantastic. I did these fixed bias rather than cathode biased to take advantage of the extra power these have. I got close to 18W before clipping and have really low THD using the thermionic labs output transformers.

Be warned new production KT120 tubes will NOT be able to be biased much higher than a KT88 due to design/manufacturing issues I uncovered. The early ones work fine and can bias at 500V and 110ma+ without signs of distress. Tungsol admitted to me that they changed how they manufactures these KT120 tubes. The newer tubes, due to a design change in getter placement, show signs of overheating/redplating as low as 500V/80ma. So you end up having to source earlier production tubes, and then hand selecting the good ones. The new KT150 tubes do the same thing. These new tube work fine in a PP amp, as they aren't biased as high as a SE amp would need to be. Also simply running a KT120 at the same bias as a KT88 doesn't sound that great and isn't worth doing.

www.skunkiedesigns.com

Skunkie Designs KT120 Monoblocks

This page covers all things related to my KT120 amp.
www.skunkiedesigns.com www.skunkiedesigns.com

KT120.png

Inside KT120.jpg
kt120_kegger-skunkie-V4.png

KT120-PSU.png
 
StepheK said:
I did a pair of KT120 mono amps based around this Kegger design and they sound fantastic. I did these fixed bias rather than cathode biased to take advantage of the extra power these have. I got close to 18W before clipping and have really low THD using the thermionic labs output transformers.

Be warned new production KT120 tubes will NOT be able to be biased much higher than a KT88 due to design/manufacturing issues I uncovered. The early ones work fine and can bias at 500V and 110ma+ without signs of distress. Tungsol admitted to me that they changed how they manufactures these KT120 tubes. The newer tubes, due to a design change in getter placement, show signs of overheating/redplating as low as 500V/80ma. So you end up having to source earlier production tubes, and then hand selecting the good ones. The new KT150 tubes do the same thing. These new tube work fine in a PP amp, as they aren't biased as high as a SE amp would need to be. Also simply running a KT120 at the same bias as a KT88 doesn't sound that great and isn't worth doing.

www.skunkiedesigns.com

Skunkie Designs KT120 Monoblocks

This page covers all things related to my KT120 amp.
www.skunkiedesigns.com www.skunkiedesigns.com

View attachment 3385073

View attachment 3385074
View attachment 3385071

View attachment 3385072
Click to expand...
hi and thanks for the tip. Fortunately, I bought all my kt120s about 5 years ago before all the nonsense with Puffin.
 
Primotubes said:
Ok. So, you don't think upping the B+ by 5 to 10% with corresponding changes in parts to handle the extra current is necessary to really take advantage of the kt120 higher output capability?
Click to expand...
Im not Mark, so take my opinion for what its worth.. (0.02 cents)
I built Stephe's KT120's (a partial pic in my avatar)...
When you say parts changes, do you mean change from self bias to fixed, PS caps, possibly output transformers etc ?
At that point, Id say may be think about taking the plunge and just build the kt120 mono-blocks...
That said, what tube type do I like to listen to the most so far ? Kt88's in UL..
 
Stephe,
Have you ever tried running the 6ej7 with g2 as anode? I tried grounding the plate, the g3, and pin6 and put b+ to g2. And I like the results. Less noice I guess. You can hear more of the decay of the notes and space around the instruments.

G2 is rated for 0.9 watts but you can exceed that some. I guess because the anode is not adding heat.
Simple to try. I would like to hear your impressions if you try it.

Thanks
 
Guess what I'm saying is that I haven't actually sat down and did all the math, looked at the tube load lines, figured up the components required - current ratings, overhead, etc. While in concept, its not all that hard, just as Stephe is saying, until you build it, you won't really know. All things are good in theory and spec sheets but bench time is critical. I can't count the times that I've sat down to build something and when I finished, it had morphed into something else based on learnings on the bench from voltage & current measurements, bias changes, feedback adjustments, the sound it produces, power supply noise, hum found, frequency response, etc. I stopped with the KT88 on my rendition of Keggar's build and haven't played beyond that in the sandbox. Looks like some others have, I applaud them for taking on the trek.

Personally, if I was trying to build a much higher output amp and stay Single Ended, I'd lean down the 805/211 patch but those aren't new roads either and the ~1000V B+ can get a little tricky. However, without that level of B+ and still attempting to squeeze a lot of power out of some SE tubes, you are starting to deal with higher currents which likely lead you right where Stephe is in what was posted above. Another path would be to go parallel SE tubes, but then you end up with a more elaborate front end to drive them. One day, I hope to have the bench time to play with stuff like this. Until then, its fun to hang out on forums like this and dream...
 
Todddo said:
Have you ever tried running the 6ej7 with g2 as anode? I tried grounding the plate, the g3, and pin6 and put b+ to g2. And I like the results. Less noice I guess. You can hear more of the decay of the notes and space around the instruments.
Click to expand...

Im lost... but up for an education lol...
grounding the plate... hmmm....

On the 6ej7 from what I see pin 6 is unused, 3&1 are cathode
G3 (pin 9) is grounded ??
You tied B+ to pin8 now making G2 the plate ??


1735657917369.png

Could you expand on how you wired it, May be a small schematic ?
Do you mean something like page 16 of the pdf below ?

Pentodes connected as triodes
http://www.audiodesignguide.com/New2A3/ETF06TS.pdf
 
You must log in or register to reply here.
Back
Top Bottom