• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

Novar Spud :: A $200 sweep tube Spud with PCB

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Member
Joined 2009
Paid Member
I have received several requests for a circuit board for a sweep tube based amplifier. Most of these center around my 6LU8 Spud, but the board for that amplifier would need significant rework to be in shape for professional fabrication. So I decided to take new look at the various options for a sweep tube Spud. The 6LU8 is a nice tube. It’s has a triode and a power pentode in one bottle. In addition to the 12-pin Compactron version, it’s available in a 9-pin Novar version labeled 6LR8. This tube is an excellent candidate for a Spud amp.

Another Novar triode-pentode combo tube is the 6KY8. The 6KY8 and 6LR8 share the same pinout. The 6GF7A is a Novar tube with a small signal triode and a power triode in one bottle. The pinout of this tube is the same as for the 6LR8 and 6KY8 – except for the screen grid pin - which, in the 6GF7A, is simply left floating. Convenient that…

After much research, experimentation, and optimization, I have come up with a design that supports all three tubes. With some creative wiring, one could use a 6LU8 as well. Note, however, that the 6GF7A runs best at a lower B+, so rolling all three tubes through is not an option without changes to the power supply. I ended up using 250 V B+ with the 6GF7A and 300 V with the 6KY8 and 6LR8.
With the 6LR8, I get about 6 W in UL mode and 4 W in triode. The 6GF7A provides about 2 W (sounds really sweet too!).

The total parts budget (including the cost of the PCB) starts at just shy of $200 (not including enclosure and shipping charges).

So without further ado, I bring to you the Novar Spud.

~Tom
 

Attachments

  • NovarSpud_AMP_1p0.png
    NovarSpud_AMP_1p0.png
    32.4 KB · Views: 4,556
  • NovarSpud_PSU_1p0.png
    NovarSpud_PSU_1p0.png
    30.4 KB · Views: 4,443
  • NovarSpud_PCB_1p0.png
    NovarSpud_PCB_1p0.png
    79.6 KB · Views: 4,312
  • NovarSpud_1p0_ProtoTop.jpg
    NovarSpud_1p0_ProtoTop.jpg
    275.2 KB · Views: 4,275
  • NovarSpud_1p0_ProtoComponent.jpg
    NovarSpud_1p0_ProtoComponent.jpg
    370.6 KB · Views: 4,237
Last edited:
Member
Joined 2009
Paid Member
The 6LU8/6LR8 is a triode-pentode Compactron. It looks like the pentode section of the EL84 has about the same max plate dissipation (12 W vs 14 W of the 6LR8) but I think that's where the similarity ends. They use different sockets as well. EL84 is a Noval tube, 6LR8 a Novar tube.

Now, you could probably take a medium mu or hugh mu triode, an EL84, and build a similar circuit.

~Tom
 
3"x5" board? That would make a nice tube amp 'mirror' to the Pass ACA diyAudio kit.

It won't fit in 3 X 5. I have a board that is nearly identical to this one. It is 4 X 6. You could physically stuff all the parts on a 3 X 5 board, but it would be cramped and heat buildup would be an issue.

I found a 6LR8 with a crack in the glass running up the entire side of the tube. I did my best to blow it up. Didn't happen. That little amp still plays even with the cracked tube.
 

Attachments

  • 500volts_signal.jpg
    500volts_signal.jpg
    351.8 KB · Views: 1,010
  • Cracked.jpg
    Cracked.jpg
    169 KB · Views: 1,143
  • PCboard.jpg
    PCboard.jpg
    318.2 KB · Views: 1,626
  • 500volts_no_signal.jpg
    500volts_no_signal.jpg
    293.7 KB · Views: 967
If any of you have comments or requests for tweaks, now would be a really good time to speak up. I plan to send the board out for fabrication early next week. :)

~Tom

Bigger PSU filter caps, snap in? No reason other than I have a few 470uF 400V snap in caps lying around, that I should probably use before they dry up.

It's not really a convincing argument, but worth a shot. :)
 
Novar sockets are rather known as magnoval (large noval) in europe.

Actually, the pins of Magnoval tubes are bigger in diameter. You can use a Novar socket for a Magnoval tube, but the pins of the socket will stretch and no longer be able to make good contact with the Novar tubes after that.

Bigger PSU filter caps, snap in?

Yep. The PSU caps are snap-in types. The footprint supports the use of caps up to 25 mm in diameter with 10 mm pin spacing. The caps I use in my build are 220 uF, 400 V, 22 mm diameter.

If the power resistors were mounted on the same side as the tubes, they could take advantage of close proximity to the metal plate/housing the PWB was mounted to in order to dissipate heat better. This might help with thermal management.

That's an interesting idea. My approach is to include some ventilation holes under the power resistors. I've seen this done in much of the HP/Agilent test equipment as well. The holes allow the resistors to breathe pretty well. The only resistors that dissipate any significant power are the resistors biasing the output stage. They dissipate about 1.4 W with the 6LR8 tubes. They're rated for 5 W...

I've measured the temperature of the various parts after a couple of hours of operation. I was mostly concerned about the caps, but they had barely gotten lukewarm. The hottest components were the 5 W resistors (no surprises there). If I recall correctly, they were running about 80~90 deg C (they're rated for 250 C!).

The board measures 3.8 x 5.8 inches (just shy of 10 x 15 cm). So 0.2 inches (5 mm) smaller on each side compared to Tubelab's board. It is possible to physically fit the parts on a smaller board, but I wanted a reasonable spacing between tubes and wanted the electrolytic caps to be a safe distance away from the hot resistors.

Note: If you open the dimensional drawing of the board in a new window and zoom in, it becomes easier to read the dimensions. Sorry for the small print...

~Tom
 
Last edited:
The Novar tubes have 0.038" diameter pins. The Magnoval uses 0.046" diameter pins. So I'm not surprised that the Magnoval sockets won't make good contact with a Novar tube (regardless of what the eBay seller says).

I bought ceramic Novar sockets from a reputable supplier - ESRC in Florida. They seem to make good contact with the tube pins. You can get the tubes there as well.

~Tom
 
If any of you have comments or requests for tweaks, now would be a really good time to speak up. I plan to send the board out for fabrication early next week. :)

~Tom

Hi Tom,

I noticed that for this new version you added some partial Schade local feedback via R3 (150Kohm), this is slightly too low based on an existing schematics nearly identical to yours from Kegger. He is using 270Kohm after many hours of listening test. Here is a link to the thread and his latest schematic.

The "$250" Amp? - 6LU8 Compactron SE, UL - AudioKarma.org Home Audio Stereo Discussion Forums

p.s. I have completed a few days ago the Kegger 6LR8 amp and it sounds great.

BR,
Eric
 

Attachments

  • 6LU8 latest fine tuned by kegger.GIF
    6LU8 latest fine tuned by kegger.GIF
    15.6 KB · Views: 1,781
Last edited:
I noticed that for this new version you added some partial Schade local feedback via R3 (150Kohm),

I tried 100k, 150k, 220k, and 330k. I found I get the best performance at 150 kΩ,

Tom copied mine ...post #2

http://www.diyaudio.com/forums/tubelab/210722-spud-what-again.html?highlight=spud+6lr8

And I copied Kegger's

http://www.diyaudio.com/forums/tubes-valves/125963-spud-anyone.html?highlight=spud+6lr8

There are experts on this forum who have told us dozens of times that this type of feedback can not be used with a triode driver. There is theory to back this statement up, but it does work quite well here. I tried several resistors and I also settled on 150K. The value seems to depend on your speakers and music choices more than anything else.
 
In addition to the Schade feedback, I tried feedback to the cathode of the output stage. That's what I used on my 6LU8 Spud, but I actually find that the amp works better with the Schade feedback.

I really like the sound with the triode coupled 6LR8 tubes. That's a rockin' li'l amp!

And yes, I allowed myself to be inspired by Tubelab, Kegger, and a few other designs on the Internetz. Thanks.

~Tom
 
Last edited:
Thanks for letting me know about the 150K vs 270Kohm, I might give it a try. The 6LR8 does make nice music ;).

p.s. To really hear the advantage of Schade feedback; omit C3 and D1 on the driver's cathode.

I also allow myself to be inspired by other.

Rgds,
Eric
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.