Question re: tube amp grounding/safety

[GXAlan]

I put in an order for a Raphaelite CS30-MKII 300B and have started this thread to share impressions and measurements along the way. In the excitement of ordering it, I failed to look at the photos carefully which is also why I'm creating this thread before its arrival and I start doing some measurements. Tagging @SIY and @amirm as our experts.

Why did I order this? Science.
There is a lot of hype about the 300B tubes being "musical." Tools from @pkane allow you to insert a distortion transfer function and to measure comparisons.

Questions to be answered once I arrives:
1. At something like 0.5W of power, will a PK Metric show a sufficient difference between a single ended 300B amplifier and a regular solid state amp? Or is the 300B tube sound all sighted bias?

2. If there is a difference, can I get a solid-state amp to match a single-ended 300B tube amp?
Tools from @pkane allow you to create distortion transfer functions and there is a general consensus that:
"If pleasant colorations are desired, you can take a transparent audio chain and add preferred colorations/distortions in software."

What "those colorations/distortions" to be added is less clear unknown. So, will it be possible to analyze this 300B amplifier and then create a transfer function that then allows the PK Metric of two musical recordings to be closer to this 300B?

Why did I pick this tube amp?
It's the right combination of attractive + value priced at <$2000 shipped via eBay. It looks like the company out of China hired a design firm from Germany to help them design the looks. I think it is a beautiful modern appearance. A quick Google search suggested that the company originally just sold audiophile transformers for hobbyists before going into their own full amplifier products, and the quality of a transformer is reported to be the an essential part of the tube sound. It also easily allows me to substitute Western Electric / RCA tubes.

Second, the company claims is that they're using a proven Western Electric circuit design. A quick glance at the internals suggested that it would be easy enough to repair/service as needed as a hobbyist. We know from HypeX DIY, that cable routing is important and it looks like it is very thoughtfully routed compared to even amplifiers from Carver.

My question, which I only recognized after putting in my order... and why I tagged @amirm and @SIY...

It looks like the IEC port is a 3 prong design. It doesn't look like the ground isn't attached to anything. The case is metal. We're dealing with lethal voltages in tube amps...

Any opinions on how I should improve product safety?

 

[SIY]

The question is, did they do the signal and PS grounding properly? I'd do a quick test by running a wire from the chassis to the GND of the IEC to see if that causes hum. If not, swap for a three pin IEC. If it does, send the shock hazard back to whence it came.

300B has about zero relation to the original tubes (which truth be told were nothing special), it's now more like a branding deal than conformance to actual specs.

 

[GXAlan]

The question is, did they do the signal and PS grounding properly? I'd do a quick test by running a wire from the chassis to the GND of the IEC to see if that causes hum. If not, swap for a three pin IEC. If it does, send the shock hazard back to whence it came.

300B has about zero relation to the original tubes (which truth be told were nothing special), it's now more like a branding deal than conformance to actual specs.

Thanks for the advice. I will update when it arrives.

I am committed to the experiment. I will make sure I source some Western Electric 300B’s for the test

 

[DVDdoug]

It looks like the IEC port is a 3 prong design. It doesn't look like the ground isn't attached to anything. The case is metal. We're dealing with lethal voltages in tube amps...

Any opinions on how I should improve product safety?

A metal case should be earth grounded (if it's not double-insulated) and I'm pretty sure it wouldn't pass UL certification. I'm not sure if it's actually "illegal". And, just the AC power coming-in is potentially lethal. It's equally dangerous with a solid state amp, unless the amp has a separate power supply.

It's safe if nothing goes wrong but the ground protects you in case of a fault that puts voltage on the case/chassis.

The case doesn't necessarily need a signal ground so it doesn't have to be prone to ground loops.

So, will it be possible to analyze this 300B amplifier and then create a transfer function that then allows the PK Metric of two musical recordings to be closer to this 300B?

It's possible but I don't know how to do it.

The advantage of software tube-emulation is that you can usually tweak it. When you buy a tube amp you're stuck with what you get and that's "dangerous" if you're looking for "tube sound" and you haven't heard that particular amp. If you don't want "tube sound" you could choose one with low distortion, but then you could save money with a solid state amp.

The disadvantage is that most of us aren't set-up to use software plug-ins.

 

[Doodski]

it looks like it is very thoughtfully routed

It has been assembled by somebody with the, "Touch." I don't know about the signal integrity of the design and layout although aesthetically it looks superb.

 

[Blumlein 88]

You can use a GFCI plug to protect yourself if there is no safety ground on that third pin. You can even buy short extensions with GFCI built in made to use with power tools. Something like this, not so pretty, but safer and you don't need to be an electrician to use it.

Tower Manufacturing Corporation 2 ft. In-Line GFCI Triple Tap Cord Automatic Reset 30438024 - The Home Depot

The 15 Amp In-Line GFCI 2 ft. Triple Tap Cord Automatic Reset is ideal for indoor and outdoor use. It features built in GFCI protection against hazardous shock from damaged cords or exposure to water.

www.homedepot.com

I notice it only has a 1 kohm input impedance. That is pretty low.

I notice they say you adjust tube noise. I wonder if they are referring to a hum adjustment?

You probably could use the 4ohm taps, use a 16 ohm power resistor tapped in the middle for half voltage and feed that into a good Solid state amp. Give you the sound of an SET and the power you need for non-horn speakers.

 

[charleski]

What "those colorations/distortions" to be added is less clear unknown. So, will it be possible to analyze this 300B amplifier and then create a transfer function that then allows the PK Metric of two musical recordings to be closer to this 300B?

I strongly suspect you will find that the distortion characteristic (balance and level of odd and even distortions) of this amp varies with output level. The easy way to check this is with the stepped level function in REW - you will see the lines corresponding to different distortion orders vary independently and cross over each other, rather than all rising in parallel as the level goes up.

This means that a fixed non-linear transfer function will only recreate the amps’s behaviour at one level. To match it fully you need a function that is ‘non-linearly non-linear’ and varies the distortion signature with level. I don’t know of any simulator that even tries to do that.

 

[SIY]

Thanks for the advice. I will update when it arrives.

I am committed to the experiment. I will make sure I source some Western Electric 300B’s for the test

The WE brand name has been licensed, as has other names from olden times (e.g., Tung Sol, Mullard...). What's made under that brand has zero relationship to original WE.

WE equipment, truth be told, was fine for 1932 PA use, but is laughably poor by later standards- and by later, I mean late 1940s and beyond. The cult of this mystifies me.

 

[SSS]

Nice looking amp. But power and distortion does not fit regular speaker boxes. Perhaps good for horns. Single ended you may get much cheaper when getting an old tube radio from the 1950ies.

 

[Sokel]

I strongly suspect you will find that the distortion characteristic (balance and level of odd and even distortions) of this amp varies with output level.

Does this applies only to tube amps or amps in general?Cause I see the same measuring different levels of a dac.

 

[SSS]

Does this applies only to tube amps or amps in general?Cause I see the same measuring different levels of a dac.

Depending on the design it is inherent in all amplifiers. Single ended tube amp has in general low negative feedback so distortion rises with the output level due to the non-linearity of the output tube. Different levels may induce different distortion harmonics caused by the active components like tubes and transistors when leaving the linear region.

 

[Sokel]

Depending on the design it is inherent in all amplifiers. Single ended tube amp has in general low negative feedback so distortion rises with the output level due to the non-linearity of the output tube. Different levels may induce different distortion harmonics caused by the active components like tubes and transistors when leaving the linear region.

Makes sense.
On the other hand that's what I get with my newbie measurements in dac,where should be linear:

Reducing DAC volume doesn't necessarily decrease bit-depth of audio data.

What you see is a line that goes down exactly with the volume, this means the absolute level of the noise/distortion remains constant. Let's see real world (attenuated at the DAC side not generator level,as a user would normally do) : -2db of attenuation 104.8 SINAD -30db of...

www.audiosciencereview.com

Reducing DAC volume doesn't necessarily decrease bit-depth of audio data.

Just for the fun of it I did a loopback of the E-MU to see,first with it's crappy analog pot from the headphone amp and then just attenuating it by windows mixer (all ASIO). at -4db at -30db crappy pot at -30 by windows mixer. Fell free to comment,suggest other measurements,etc...

www.audiosciencereview.com

 

[computer-audiophile]

It has been assembled by somebody with the, "Touch." I don't know about the signal integrity of the design and layout although aesthetically it looks superb.

The amplifiers that are made in China under the label 'Raphaelite' are above average and 'classically' made compared to the overall range that can be found on AliExpress, for example. They also make good output transformers. I find Line Magnetic a little better, but also more expensive. I have some experience with this type of equipment.

I think it's great that the Chinese have taken up the tube theme and continue to cultivate it. The market is also very interesting for DIY components, I think. The tubes from Shuguang or Linlai, e.g. the 300B types, have also found recognition among connoisseurs.

 

[charleski]

Does this applies only to tube amps or amps in general?Cause I see the same measuring different levels of a dac.

Yes, you can usually find similar behaviour with solid-state amplifiers. The difference is that a decent SS amp has such low distortion that the way the characteristic varies with level isn't really noticeable. I posted a gif animation to demonstrate this in my measurements of an old tubed headphone amp here.

 

[fpitas]

This SS amp might give you the same sort of sound as a SET:

https://www.firstwatt.com/pdf/prod_sit3_man.pdf

Most listening impressions I've seen say it's not a bad approximation, although somewhat cleaner sounding.

 

[Zapper]

I strongly suspect you will find that the distortion characteristic (balance and level of odd and even distortions) of this amp varies with output level. The easy way to check this is with the stepped level function in REW - you will see the lines corresponding to different distortion orders vary independently and cross over each other, rather than all rising in parallel as the level goes up.

This means that a fixed non-linear transfer function will only recreate the amps’s behaviour at one level. To match it fully you need a function that is ‘non-linearly non-linear’ and varies the distortion signature with level. I don’t know of any simulator that even tries to do that.

Agreed. It can get even more complicated. Because of power supply droop, capacitive coupling to the grid, and capacitive bypass of the cathode bias, the DC bias point of the tube can shift dynamically as a complex function of the signal history with several time constants. So now you have 'history dependent non-linear non-linear'.

Guitar amp designers and modders take advantage of these history dependent effects to alter guitar tone during attack vs sustain, for example. More supply droop can give a punchier attack relative to sustain. The shifting bias can be tweaked to give a cleaner or more distorted sustain, etc. There has been a lot of work on digital emulation of tube amp sound and many products based on it for musicians.

 

[computer-audiophile]

This SS amp might give you the same sort of sound as a SET:

https://www.firstwatt.com/pdf/prod_sit3_man.pdf

Most listening impressions I've seen say it's not a bad approximation, although somewhat cleaner sounding.

I find 'sort of sound' a funny generalization in context. After all, you didn't say they sound exactly the same.

 

[fpitas]

I find 'sort of sound' a funny generalization in context. After all, you didn't say they sound the same.

Well, even two different tube amps that distort badly enough to be audible won't sound the same, for the reasons described in previous posts.

 

[fpitas]

The question is, did they do the signal and PS grounding properly? I'd do a quick test by running a wire from the chassis to the GND of the IEC to see if that causes hum. If not, swap for a three pin IEC. If it does, send the shock hazard back to whence it came.

300B has about zero relation to the original tubes (which truth be told were nothing special), it's now more like a branding deal than conformance to actual specs.

And depending on who you ask, a NOS type 45 is the cat's meow. NOS 2A3s also get lots of love. The shape of the plate is seen as important etc.

 

[computer-audiophile]

It is not that difficult to make these classic electron tubes new. I think it's worth trying out for yourself what the Chinese have been offering for some time. In the past, I only used NOS tubes. Today I think NOS is more something for collectors and the prices are crazy. By the way, I agree with many designers that there are many other factors that influence the sound of a power amp more than the type of tube itself.