30+ watt amp the size of a pedal!

[SoulFetish]

Quilter's been making quality solid state power amps for a long time. I respect the dude. He's not the first to take advantage of class D output stages for guitar amps, but he's creatively packing some decent power in small enclosures.
David Bering has some really interesting techniques using switching in tube output stages. He calls it Zero-Hysterysis Output Transformerless... wait, was he trying to get cute with that name? I just realized that it kind of rhymes and it just became bit more lame. Ugh. I'm gonna' let it slide this time, because it otherwise is pretty interesting.
Tube amplifiers for high-end audio by The David Berning Company
http://davidberning.com/images/paten...ent5612646.pdf

I wonder if they drive the speaker directly with a PWM signal counting on the speaker to be an LPF?

[loudthud]

I wonder if they drive the speaker directly with a PWM signal counting on the speaker to be an LPF?

Like R.G. says, there are no regs in the US about how dangerous your product can be, but there are regs about how much noise your product can radiate and inject into the mains. An unfiltered PWM signal on a speaker wire would be a huge radiator.

Where does that come from?

I was speaking of the "wall wart" type AC transformers that hang off of wall outlets or sometimes have a line cord much like laptop power supplies, but are not inside a product. Back in the early 90's I was looking at product offerings and 24VAC 40VA seemed to be the limit of what was available. The vast majority do not have the third wire safety ground (unless you pay extra). The safety ground helps keep the hum down when you try to make a high gain guitar amp using a wall wart type transformer. AC Voltage regulation is poor but I think it makes the amp sound and feel more like a tube amp. For a SMPS I'm tempted to install a resistor inline with the DC to deregulate the supply.

[SoulFetish]

Like R.G. says, there are no regs in the US about how dangerous your product can be, but there are regs about how much noise your product can radiate and inject into the mains. An unfiltered PWM signal on a speaker wire would be a huge radiator.

Also, wouldn't the PWM magnetic field generate high voltage spikes on the speaker coil?

[nickb]

Quilter's been making quality solid state power amps for a long time. I respect the dude. He's not the first to take advantage of class D output stages for guitar amps, but he's creatively packing some decent power in small enclosures.
David Bering has some really interesting techniques using switching in tube output stages. He calls it Zero-Hysterysis Output Transformerless... wait, was he trying to get cute with that name? I just realized that it kind of rhymes and it just became bit more lame. Ugh. I'm gonna' let it slide this time, because it otherwise is pretty interesting.
Tube amplifiers for high-end audio by The David Berning Company
http://davidberning.com/images/paten...ent5612646.pdf

I really don't know how he can justify calling it "transformerless" when it uses two where one used to work. The audio frequency one has been replaced by a ferrite one so it will be smaller and lighter with better HF performance, all good attributes, but at the expense of complexity.

I wonder if they drive the speaker directly with a PWM signal counting on the speaker to be an LPF?

The switching frequency in Class D is going up, now over 2MHz meaning the inductor is tiny.

[nickb]

Also, wouldn't the PWM magnetic field generate high voltage spikes on the speaker coil?

Most (all?) of these FETs have a built-in source-drain diode which will clamp any spikes during the off-time from destroying the FETS to the power rails. Sometimes they add external diodes across the output FETs too.

[printer2]

I want to play with Fets yet but for now I plan on running a Marshall preamp into a switching amp, TPA3118 BTL.



The tubes run off a boost converter, the PA and the tubes run off a 28V laptop brick.



Picked out the lightest piece of pine board I could find, cab the size of a 5E3. Speaker fairly light weight Yamaha.



I have a Yamaha that is more efficient and I like the sound a little better but it has a cast frame and a heftier magnet.

[Zozobra]

I want to play with Fets yet but for now I plan on running a Marshall preamp into a switching amp, TPA3118 BTL.



The tubes run off a boost converter, the PA and the tubes run off a 28V laptop brick.



Picked out the lightest piece of pine board I could find, cab the size of a 5E3. Speaker fairly light weight Yamaha.



I have a Yamaha that is more efficient and I like the sound a little better but it has a cast frame and a heftier magnet.

I'll be following this with interest

[printer2]

I'm curious as to what it will sound like. Probably won't be doing it for another month or so as I am building a Tele and I have a couple of people waiting for their acoustic guitars at the moment.

[Zozobra]

I have my ZVS board and a DC-DC converter for the heaters on order so I may get to start fiddling fairly soon. The laptop PSU I was planning on using doesn't have an earth bond to the plug though so I need to think about how to proceed with that from a safety POV. Would connecting the chassis to the DC negative be sufficient? Seems a little flaky.

[loudthud]

The laptop PSU I was planning on using doesn't have an earth bond to the plug though so I need to think about how to proceed with that from a safety POV. Would connecting the chassis to the DC negative be sufficient? Seems a little flaky.

There will likely be a small current that wants to flow between the output (either side) and Earth ground. It may simply be stray capacitance to one side of the Mains or possibly noise from the switching inverter. Without a safety ground, this current wants to flow from the guitar strings through you to that damp floor you are standing on or a properly grounded microphone. A DVM (set to AC Volts) or oscilloscope will usually show a substantial Voltage between Earth ground and either side of the DC output.

Not a good situation. This noise will probably find it's way into the audio path, another thing to avoid. There is also the possibility that a component failure inside the PSU could allow Mains current to find it's way to the output posing a lethal hazard. To be safe, you need a ground.

In the US some products have a two wire polarized line cord. This allows you to only insert the Mains cord one way into the wall outlet. Since one side of the Mains (called neutral) is grounded at the breaker panel (again, in the USA) it can sometimes be used to minimize the current, but not totally eliminate it in all situations. I don't know if this is the case in your country.

This can not be over emphasized: TO BE SAFE YOU MUST HAVE A GROUND !

[printer2]

I have nothing against having a ground, I am just curious about the concern about the safety aspect. Every PSU that I took apart (hey, I have to figure out how things work) the circuit attached to the mains and the secondary side is isolated from each other and only is connected through the magnetic circuit. One side resides on its area of the circuit board and the other on its. I guess you can conceive of a component failure and the board or the transformer melts down and a path to the mains is produced, no different than a regular transformer in our amps frying and shorting out between the primary and secondary.

[nickb]

I have nothing against having a ground, I am just curious about the concern about the safety aspect. Every PSU that I took apart (hey, I have to figure out how things work) the circuit attached to the mains and the secondary side is isolated from each other and only is connected through the magnetic circuit. One side resides on its area of the circuit board and the other on its. I guess you can conceive of a component failure and the board or the transformer melts down and a path to the mains is produced, no different than a regular transformer in our amps frying and shorting out between the primary and secondary.

The design of the transformers differs between grounded (Class I) systems and double insulated (i.e no grounded case class II). The basic idea as I understand it is to have two levels of protection. In a class I system the transformer will have basic insulation plus a metal screen between the primary and secondary. The screen is connected to safety earth and so protects against insulation failure. For a class II system the insulation is reinforced to provide the second level of protection. This is rather simplified but hopefully you get the general idea.

[Zozobra]

Yeah, I figured it was a no-go from a safety POV. The AC inlet to the brick is 2 pin and no ground passthrough so I'll not be using it to drive my ZVS. Maybe I'll use it for a class D poweramp or something similar...

[uneumann]

Yeah, I figured it was a no-go from a safety POV. The AC inlet to the brick is 2 pin and no ground passthrough so I'll not be using it to drive my ZVS. Maybe I'll use it for a class D poweramp or something similar...

There are two issues here that seem to be mixed in the discussion. One is whether your chassis has an earth ground and the other is whether your PS brick has an earth ground. IMO - disclosure: I'm not a code/UL safety expert - having an earth ground and a 2pin PS are not incompatible.

Use a 3 prong cord to the chassis and make an earth ground to the chassis. Then - after the switches and fuse - wire the 2 pin PS brick as you would wire any 2-wire transformer primary. The output of the brick will need a connection to signal and chassis ground - as any DC supply does. The idea is to treat the brick as you would any transformer/rectifier/filter DC supply.

[printer2]

Use a 3 prong cord to the chassis and make an earth ground to the chassis. Then - after the switches and fuse - wire the 2 pin PS brick as you would wire any 2-wire transformer primary. The output of the brick will need a connection to signal and chassis ground - as any DC supply does. The idea is to treat the brick as you would any transformer/rectifier/filter DC supply.

Was thinking about it and came up with the above as a possible solution, glad to see I am not the only one.