RobRob Stereo

DestroXVII

TDPRI Member
Joined
Jun 19, 2022
Posts
29
Age
41
Location
Massachusetts
Hi everyone,

I’ve built a RobRob Stereo 6V6 amp that is destined for a ‘34 Zenith cathedral cabinet. It works, but it puts out low power with the outputs grounded.

Leaving the outputs ungrounded yields apparent full power (2-3 watts-ish?) but no high end—very heavy bass.

Grounding the outputs yields excellent, balanced sound but very little power (I estimate 1/4 watt). Dumping a lot of audio signal into the cake pan, I guess?

I’ve improved my grounding scheme, but it’s not a perfect star system due to space constraints.

I’ve also discovered that the Antek 280v PT puts out 406vdc unloaded but only 324 with the tubes in place.

Plate voltage at the drivers is very close at about 268, while the screens and plates of the power tubes are low at 285 and 288, respectively.

Any ideas? I’ve checked the circuit several times, and it’s wired according to the schematic.

Thanks in advance,
Jonathan
 

woodandglue

TDPRI Member
Joined
Apr 14, 2020
Posts
50
Location
right here
Hey there, I don't have an answer for you (I like to build and struggle to diagnose), but posting a schematic and pictures of your actual build will usually result in better, quicker responses. I hope that helps!
 

2L man

Tele-Afflicted
Joined
Nov 23, 2020
Posts
1,547
Age
62
Location
Finland
Please explain what kind grounding you mean? It should effect output power only thru Negative Feedback.

If amp has no NFB it is good to ground output when othervise OT leakage might elevate the voltage high. Also a resistor grounds secondary enough.
 

DestroXVII

TDPRI Member
Joined
Jun 19, 2022
Posts
29
Age
41
Location
Massachusetts
Hi,

Thanks for responding. It’s the simpler circuit with no preamp:


Antek support said I could increase filtering or decrease plate current. I wonder if switching to the 10k ohm tap would make a difference. Currently, the outputs are on 5k primaries.

PS - can put pics up tomorrow.
 

DestroXVII

TDPRI Member
Joined
Jun 19, 2022
Posts
29
Age
41
Location
Massachusetts
2L man, the ground I’m referring to is for the speaker outputs. Some say that OTs don’t care whether or not their secondaries are grounded, but I have not found that to be the case.

In this case, it is definitely not the case.
 

andrewRneumann

Friend of Leo's
Joined
Mar 22, 2020
Posts
2,023
Location
Cincinnati, OH, USA
2L man, the ground I’m referring to is for the speaker outputs. Some say that OTs don’t care whether or not their secondaries are grounded, but I have not found that to be the case.

In this case, it is definitely not the case.

While troubleshooting, disconnect NFB until this problem is resolved. If that solves it, then we know something is wrong with NFB.

If you post a schematic, layout and gut shots I bet you will get more responses.
 

DestroXVII

TDPRI Member
Joined
Jun 19, 2022
Posts
29
Age
41
Location
Massachusetts
While troubleshooting, disconnect NFB until this problem is resolved. If that solves it, then we know something is wrong with NFB.

If you post a schematic, layout and gut shots I bet you will get more responses.
Thanks, Andrew. The layout and schematic are linked in one of my response posts. It’s on Rob Robinette’s site. Not at home right now, but I will post gut shots when I can.
 

DestroXVII

TDPRI Member
Joined
Jun 19, 2022
Posts
29
Age
41
Location
Massachusetts
13A1DF1E-C563-44FD-8539-BB6A26833A6E.png
93CCFE19-AB8E-48E5-8766-46EA34C21ACC.png
 

andrewRneumann

Friend of Leo's
Joined
Mar 22, 2020
Posts
2,023
Location
Cincinnati, OH, USA
Thanks, Andrew. The layout and schematic are linked in one of my response posts. It’s on Rob Robinette’s site. Not at home right now, but I will post gut shots when I can.

Thank you. As someone who likes to help out, I really like the courtesy when someone just posts the schematic and layout they are working from instead of making everyone go find it. Rob’s site, amazing as it is, is not the easiest to navigate. Some of the those pages get pretty long! It’s like reading the Torah from a scroll. 😉

This is such a simple amp, we should be able to get it squared away. Did you go with binding posts or 1/4” jacks?
 

andrewRneumann

Friend of Leo's
Joined
Mar 22, 2020
Posts
2,023
Location
Cincinnati, OH, USA
When you get eyes on it again, check the wiring of the NFB circuits. There is a backwards cap in Rob’s layout and it is not clear which wire of the NFB is shield and which is signal.

Personally don’t feel shielded wire is necessary for NFB. OMMV.
 

2L man

Tele-Afflicted
Joined
Nov 23, 2020
Posts
1,547
Age
62
Location
Finland
2L man, the ground I’m referring to is for the speaker outputs. Some say that OTs don’t care whether or not their secondaries are grounded, but I have not found that to be the case.

In this case, it is definitely not the case.
NFB needs two wires to operate. Agent technology use Chassis as one.

That layout looks fine and NFB use wires and if banana jacks are insulated?

Only what I wonder is the purpose for ground signs on bith side of the top electrolyt but there are no wires so obviously they are not used?
 
Last edited:

DestroXVII

TDPRI Member
Joined
Jun 19, 2022
Posts
29
Age
41
Location
Massachusetts
NFB needs two wires to operate. Agent technology use Chassis as one.

That layout looks fine and NFB use wires and if banana jacks are insulated?

Only what I wonder is the purpose for ground signs on bith side of the top electrolyt but there are no wires so obviously they are not used?
Bananas are insulated. The ground symbols on the terminal strips aren’t used. Layout shows them to make note that they are there.
 

andrewRneumann

Friend of Leo's
Joined
Mar 22, 2020
Posts
2,023
Location
Cincinnati, OH, USA
Here’s what I’m talking about in post #11

1656245691372.jpeg


Now, for the issue of lack of high end. I would consider finding a lower value volume pot. I don’t know if you can find a dual gang 10K audio pot, but I think that would really help. If you can’t find a dual gang, do you have room for individual L and R volume pots?

EDIT: If you are into the big 24mm pots, Mouser has this one: Alpha 10K DG pot. They have many other sizes and styles to choose from.
 
Last edited:

robrob

Poster Extraordinaire
Ad Free Member
Joined
Dec 29, 2012
Posts
8,625
Location
United States
The current layout is V1.9a which has the corrected cap polarity. If you see something earlier (like the V1.7 shown here) on my website you need to do a hard refresh. In the Chrome browser that is done by holding down the shift key when you click on the page refresh button. The old version shown here may be caused by tdpri caching images. I have to change the image name to get around the old, cached images.

The NFB shield is connected to a terminal that is directly connected to chassis ground.
 

DestroXVII

TDPRI Member
Joined
Jun 19, 2022
Posts
29
Age
41
Location
Massachusetts
Thanks for all the help, guys.

I wired based on the schematic, so my driver bypass caps are correctly oriented. The NFB shields connect directly to ground, too.

Because I’m (literally) using an 11x7 cake pan that fits into my cathedral radio, I couldn’t follow Rob’s proposed layout. The amp sounds great, and there’s negligible noise—my only issue is the low power.

My vaguely educated guess at the cause of the problem is the 288v I’m getting at the plates and 324 at the OTs with the 280v-rated Antek toroidal.

Antek said the 363 B+ should be be 370 at least with 390 being on the higher end. They recommended more filtering or decreasing current. 363, according to them, is ok for the PT, but it’ll run a bit hotter.

Right now, I’m using the yellow secondary wire of the Hammond 125DSEs for 5k on the 6v6s. I could use another tap for 10k, which is at the top end of the recommended 6v6 plate impedance.

Would this not result in an current decrease and a voltage increase?

Thanks!

PS - I know it’s a rat’s nest. Snicker privately, please.
 

Attachments

  • 339A0A9F-907C-49FC-9452-7855433593BE.jpeg
    339A0A9F-907C-49FC-9452-7855433593BE.jpeg
    270.5 KB · Views: 31
  • E5CF1E95-0FEC-4CBC-8324-5E0B061FD0E7.jpeg
    E5CF1E95-0FEC-4CBC-8324-5E0B061FD0E7.jpeg
    218.6 KB · Views: 31
  • 05D3766B-4EEF-44D9-A841-3C6E60791688.jpeg
    05D3766B-4EEF-44D9-A841-3C6E60791688.jpeg
    234.7 KB · Views: 31
Last edited:

DestroXVII

TDPRI Member
Joined
Jun 19, 2022
Posts
29
Age
41
Location
Massachusetts
The current layout is V1.9a which has the corrected cap polarity. If you see something earlier (like the V1.7 shown here) on my website you need to do a hard refresh. In the Chrome browser that is done by holding down the shift key when you click on the page refresh button. The old version shown here may be caused by tdpri caching images. I have to change the image name to get around the old, cached images.

The NFB shield is connected to a terminal that is directly connected to chassis ground.
Rob,

As a side note, I successfully built your JCM 6v6 into a 1929 A-K breadbox. It’s a junky salvage and not 100% done, but I’ll send pics if you like.

Still tuning and plan to add a cold clipper rotary switch, but it sounds pretty ballsy.

Thanks for the circuit!
 

andrewRneumann

Friend of Leo's
Joined
Mar 22, 2020
Posts
2,023
Location
Cincinnati, OH, USA
Thanks for all the help, guys.

I wired based on the schematic, so my driver bypass caps are correctly oriented. The NFB shields connect directly to ground, too.

Because I’m (literally) using an 11x7 cake pan that fits into my cathedral radio, I couldn’t follow Rob’s proposed layout. The amp sounds great, and there’s negligible noise—my only issue is the low power.

My vaguely educated guess at the cause of the problem is the 288v I’m getting at the plates and 324 at the OTs with the 280v-rated Antek toroidal.

Antek said the 363 B+ should be be 370 at least with 390 being on the higher end. They recommended more filtering or decreasing current. 363, according to them, is ok for the PT, but it’ll run a bit hotter.

Right now, I’m using the yellow secondary wire of the Hammond 125DSEs for 5k on the 6v6s. I could use another tap for 10k, which is at the top end of the recommended 6v6 plate impedance.

Would this not result in an current decrease and a voltage increase?

Thanks!

PS - I know it’s a rat’s nest. Snicker privately, please.

Hey that's a cool project.

IMO, 324V on the plates is plenty to give you sufficient volume. Definitely more than a 1/4W!

Back to your original problem. The signal should not cut out when one and only one tap of the OT secondary is grounded. This indicates a problem if you cannot ground the secondary. If you disconnect the grounds and measure resistance from each binding post to ground, you should get essentially 1.9kΩ. (This assumes NFB is connected. With no NFB, you should read OL--open circuit.) Can you verify that please?
 

andrewRneumann

Friend of Leo's
Joined
Mar 22, 2020
Posts
2,023
Location
Cincinnati, OH, USA
Are you absolutely sure that the shield of the NFB is in no way making contact with the NFB signal conductor? Check both ends and make sure that shield is well clear of signal. Did you try it with NFB disconnected at the binding post end?
 




New Posts

Top