Silent Trainwreck!

[Noobytube]

Hi all,
I’m new here and appreciating all u guys and this amazing resource! I’m currently collecting parts for my first Express clone build, I will be using Pacific tranny’s from RJ and Mallory OD’s from Rooster, I’m not planning to build as a hardcore purist, I’m interested in building the quietest/hiss minimal clean to mean Trainwreckesque amp I can manage. Like many I was blown away with GK’s (geetarpickers) express video on YouTube, the range and tone of his amp speaks for itself, the 59LP and playing don’t hurt either! The only small drawback that I heard was the amount of hiss the amp displayed, I’m well aware that this is an unavoidable side effect of high gain amps.
That said, I’m planning on/or considering applying the following processes in my build to help bring down the hissing beast as much as possible, I realize and look forward to the reality that this amp requires a virtually perfect lead dress and a lot of tweaking to dial it in up to the coveted edge of disaster, exactly how the great Ken Fisher did it! I hope I can come even remotely close.
Please feel free to shoot down anything on, and PLEASE add any help to this list! Hopefully others will find it useful as well.
Much appreciated!

In addition to normal Express clone build practices and in no particular order....
1. Use wirewound mills anywhere possible
2. Wire input stage p to p in an effort to concentrically locate all the v1 control components.
3. Use conductive plastic pots
4. Foil side of caps toward grounds and previous stages
5. Considering a metal shield between psu board and ripple caps/pre board
6. Following the star ground/cascading priority of Keen article
7. Shuerter iec mains input filter.
8. As quiet as achievable possible v1 ax7
9. Suspend and space the chassis off the cab floor and add ventilation holes under/ above boards as hotter=more thermal noise
And ?..........

Thanx!
Anthony

[romberg]

The hiss involved with high gain tube guitar amplifiers is something you can try and minimize. But it is a battle you can't do much to totally win. See:

http://www.aikenamps.com/index.php/resi ... -it-matter

And I highly recommend Merlin Blencowe's guitar preamp book which has a whole section on noise.

http://www.valvewizard.co.uk/

Most of the big things to reduce hiss have already been done by Ken in the design (use of resistors is minimized). I think most of your ideas such as pots and wire wound resistors sound like good ideas. But beware of the adding in quadrature problem mentioned by the above two authors.

Because the noise adds in quadrature [total noise = sqrt( N1 * N1 + N2 * N2 +...)], if the noise is dominated by one source then it will make little difference reducing any other source of noise. You will just be making your build more difficult for very little if any benefit. This is compounded by the fact that guitar amps that sounds good overdriven, by design compress the signal to noise ratio. Your guitar signal smashes up against the wall when the tubes clip while the noise is still amplified by all the gain the circuit has. This the signal to noise ratio is reduced.

One thing you could do with an express, is use the extra triode as a parallel input stage. A parallel stage will produce less noise because less current is required for a pair of triodes to produce the same amount of gain. Merlin mentions this in his book. He also mentions that the gain in signal to noise ratio is also pretty much swamped out by the grid stopper you are going to want/need on V1 anyway. But as long as you have an extra triode...

I am fairly sure that trying to reduce the chassis temperature by going overboard and trying to create airflow is another one of these areas where the benefit may not even be measurable. I think your #5 and #9 ideas are mostly creating more difficulty where none is needed. They would help with hum and buzz coming from the power supply. And this is not a common problem with this circuit. The hiss is mostly about the grid restistors and the current flowing in the tube. Again, Merlin spells this out in quite some detail.

[geetarpicker]

Glen here! Probably the best way to reduce the hiss for home use is to simply use a power attenuator. These amps in my experience have less hiss than most amps that (very importantly) have similar gain. Cranked up in the room with a 4x12 these amps can hit between 110-120db. That said attenuate it down maybe 12db and it will still be loud around the house, but the noise is then 12db down also. Attenuated to bedroom levels the hiss will be effectively GONE. Personally I WOULD NOT suggest getting too experimental on your first wreck build as it may bite you with problems. GK

[Winder]

+1 on the attenuator.

[Noobytube]

Thanx guys, good stuff. Any recommendations on attenuators?

[Winder]

I own a THD HotPlate and a resistive attenuator and prefer the resistive. It's a 200W "Output Tamer" I got from an eBay seller. I set the HotPlate to load and use it as a dummy for setup and testing. There are schematics on TAG for the AirBrake. I plan to build on at some point. There are a ton of choices out there, so depends on how many options you want (line out, DI, etc.). I just want to knock the volume to a tolerable level through a 4x12 but still move a bit of air.

[Roe]

The alex slash scumback attenuator works well as does the fryette power station

[Noobytube]

The hiss involved with high gain tube guitar amplifiers is something you can try and minimize. But it is a battle you can't do much to totally win. See:

http://www.aikenamps.com/index.php/resi ... -it-matter

And I highly recommend Merlin Blencowe's guitar preamp book which has a whole section on noise.

http://www.valvewizard.co.uk/

Most of the big things to reduce hiss have already been done by Ken in the design (use of resistors is minimized). I think most of your ideas such as pots and wire wound resistors sound like good ideas. But beware of the adding in quadrature problem mentioned by the above two authors.

Because the noise adds in quadrature [total noise = sqrt( N1 * N1 + N2 * N2 +...)], if the noise is dominated by one source then it will make little difference reducing any other source of noise. You will just be making your build more difficult for very little if any benefit. This is compounded by the fact that guitar amps that sounds good overdriven, by design compress the signal to noise ratio. Your guitar signal smashes up against the wall when the tubes clip while the noise is still amplified by all the gain the circuit has. This the signal to noise ratio is reduced.

One thing you could do with an express, is use the extra triode as a parallel input stage. A parallel stage will produce less noise because less current is required for a pair of triodes to produce the same amount of gain. Merlin mentions this in his book. He also mentions that the gain in signal to noise ratio is also pretty much swamped out by the grid stopper you are going to want/need on V1 anyway. But as long as you have an extra triode...

I am fairly sure that trying to reduce the chassis temperature by going overboard and trying to create airflow is another one of these areas where the benefit may not even be measurable. I think your #5 and #9 ideas are mostly creating more difficulty where none is needed. They would help with hum and buzz coming from the power supply. And this is not a common problem with this circuit. The hiss is mostly about the grid restistors and the current flowing in the tube. Again, Merlin spells this out in quite some detail.

Great thanx!
I was planning on trying a 5 star ge 6072 red ink with perfectly matched triodes and next to no noise in parallel in the input stage, I’m not ever going to need the 12ax7 gain the express is capable of in its extremes..., so I figured I’d play a bit with that lower gain tube, any advice on implementing? I assume just share the same plate resistor and cathode bypass? Of corse I will check regardless

Thanx again

[romberg]

Merlin covers parallel gain stages in his book and you can search for other information on the web. In a nutshell, to get the same gain and bias point on a parallel stage as you have on a single valve, you halve the shared plate and cathode resistors and double the cathode bypass cap (for the same RC constant). You need to share the plate resistor. But you can use separate cathode resistors (in which case you don't 1/2 them) for different tones by biasing the tubes differently.

I would not do this until *after* you have built the circuit as "stock". You may in fact find that you do want all the gain an express has. So, start with stock and then modify.

Mike