Calling all Techs --Crazy NAT tube amplifier

[Roy]

I am an electronics Tech with far more experience in solid state
and digital circuits, than with vacuum tube equipment.
Customer brought in an
NAT model # "Single" It is a high-end SET Stereo Integrated amp.

I gave up on it recently. Each channel uses a single #805 Triode
driven by a 6H1N dual section tube.
I have NEVER seen any tube amp with negative 530 volts on the GRID
of the #805 tubes and the PLATE (which shows negative 20 volts) connected thru the
output transformer PRIMARY to CHASSIS/CIRCUIT GROUND.
On google images I don't see ANY schematics showing this scheme
or that any manufacturer has ever done this.
I have always seen a high POSITIVE voltage on the plates of tube amps.
I don't know if a tube amp can actually work with the plates of
the output tubes referenced to ground.

Both channels only produce about 3 watts RMS into 8 ohms.
This is a TWIN AMP / completely discrete design.
The only channel commonality is the line cord, the power switch and power transformer.
Manufacturer (in Serbia) promotes this as a 50 watts per channel amp.

There is a huge amount of sine wave signal on each output tube grid but
very weak at the plate. I don't see any feedback loop problem.
My opinion is the amp was made this way.....a mistaken design.
I have some dis-jointed schematics of this amp but they
are too simplified and hugely unhelpful.

The amp does play thru speakers ....so the customer did not know
the output was so weak. He only brought it to me to replace
frayed anode wires and a couple of broken RCA jacks at the CD input.

I would like to refer him to someone local in South Florida with more tube
experience than I have and that he can bring this 110 pound amp to.
Customer will spend good money to repair, since this amp is worth
something like 20,000 dollars.
I can't imagine him shipping this extremely heavy thing anywhere.

Thank you for your time ! Anyone in South Florida interested in tackling this
please contact me at : roy-lakes@comcast.net

=============================================

[Enzo]

as far as the tube is concerned, there is no difference between +500v on the plate and a grounded cathode, and a grounded plate and -500v on the cathode. Either way the tube only sees it has a plate 500v more positive than its cathode.

Would that be unconventional? Sure, but otherwise not an issue.


Imagine a TTL chip, usually +5 on the top and ground on the bottom, but just as happy with ground on top and -5 on the bottom. We have some amps that have op amps running between ground and -30v instead of between +/-15v.


If the customer used to be happy with it, are we sure he will be happy if we "fix" this issue?

[J M Fahey]

Agree and add: Email NAT in Serbia, comment on the 3W RMS output and ask for schematic and setup instructions to leave it as good as new.

[Randall]

"The amp does play thru speakers ....so the customer did not know
the output was so weak. He only brought it to me to replace
frayed anode wires and a couple of broken RCA jacks at the CD input."

So I'm thinking, why not do the repairs he asked for and get it out of your shop? The rest is on him.

[Roy]

Thank you Enzo for your wisdom. I figured the output tube wouldn't care about the polarity
as long as the plate was far less negative than the grid is but I have never seen anyone configure an output tube that way.
I cannot do anything more with this amp, myself....because I spent a ridiculous amount of time searching
for the reason two totally discrete channels behave the same, and the schematics are so simplified and worthless.
Just trying to see if anyone local here is willing to have a shot at it to help this customer since I can do no more.
Again, my thanks for your comments.

[Roy]

Thanks for your comment. I have been in touch with NAT. They are of no help. There is another tech I heard of working on other NAT amps
here in USA that is not having any success at all repairing them. The schematics NAT gave are only little disconnected pieces
of circuits. Huge amounts of circuitry are missing from these paltry documents. They refuse any other help. They seem to be
trying to protect their intellectual property from copying.
Long ago I gave up on the power issue and wrapped this up for the customer to pick up when he returns from travel.
Long go I replaced the jacks and the anode wires as he had requested. So I am done with it.

[Roy]

Thanks Randall --
It never "was working fine" in terms of producing the rated 50 watts from either absolutely discrete channel.
The very few clean watts it produces thru speakers makes it appear that it is.
I don't think two discrete channels could be this way unless
1. It came that way from the factory or 2. Someone modified or made an improper change in BOTH Channels.
I haven't been able to ask the owner of the amp if it has ever been worked on by anyone since he bought it.

Long ago I gave up on the power issue and wrapped this up for the customer to pick up when he returns from travel.
Long ago I replaced the jacks and the anode wires as he had requested. That is all I could do. So I am done with it.
But perhaps someone other than me wants to have a look at it for this customer.

[eschertron]

I'm just talking out my blowhole, but I might expect a super-high-fidelity amp using an SE configuration to only produce a fraction of the rated output power of the tubes, in order to stay in the 0.001%-ish THD linear range. A guitarist would want to hear the 5%+ distortion as the tube nears its limit. An audiophile... not so much. Besides, 3W per channel is plenty loud for the average living room

edit: so I vote for option 1. It came that way from the factory because the designer wants it that way. But you can't sell to the masses with a "3W per channel" claim.

edit 2: just visiting a random maker site, they advertise 1% THD at 48W at exactly 1kHz. No mention of what the distortion is at the 10Hz they list as the low end of the bandwidth.

edit 3: suggest to the customer to get another pair of 805s. If they're weak then the 'high grid drive voltage' and 'weak anode voltage' makes sense.

edit 4: as far as design goes, most of the random amp makers I visited used 6L6s or equivalent as drivers, not winky little EC88s. Maybe that makes a difference.

[g1]

Another option: there is something weird about the circuit that doesn't like the scope or test set-up.
Suggest measuring with multimeter on AC volts across speaker or floating load resistor (connected to nothing else other than meter).

[Mick Bailey]

I have always seen a high POSITIVE voltage on the plates of tube amps.

There are all kinds of novel tube configurations within the world of Hi-Fi that are outside of what we expect. For example, there's an 'inverted triode' connection used with transmitting triodes. The plate is held at about 500v NEGATIVE, the signal fed to the cathode and the output taken directly off the grid with no output transformer. With an 805 driven with a 100v signal you get about 4W. The claim is excellent linearity (so no NFB needed) and low distortion. Alternatively, the plate has a HT negative bias of several hundred volts, the cathode also has a small negative bias and the input capacitor-coupled to the plate. Again, the grid is used for the output.

There are some issues with the 805 distorting in regular class A configurations at higher output levels and sometimes they're operated with the grid in a positive region (A2) to overcome this. Maybe the amp you have is an unconventional design intended to overcome the shortcomings of a regular class A 805 amp.

[Roy]

NAT Amplifier issue -- Problem Solved

Thanks to everyone who responded with comments. When I first posted ...being a newbie here and having never posted to any online forum before....
I wrongly posted to two different forums on this site. Enzo and Randall pointed out to me that dual posting wasn't necessary.
Anyway...I solved the problem with measuring the power output of this NAT Single Mkll amplifier. Please see my other forum listing
for the NAT amplifier if you would care to see the resolution.

[g1]

See here for that resolution: https://music-electronics-forum.com/...l=1#post538839