THR100 testing

Almach

Measured a few things on my THR100 today



There's a permanent 192Khz waveform on the output of around 300mv pk/pk. Probably the switched mode power supply.

Almach

192Khz sine wave on output.

Almach

Input at max output



That's around 140W ( handbook says 40W ! )

Almach

Nearly 72 V pk/pk flat out using EL34/Clean/16 Ohm into a 11.7 Ohm load. Red is 1Khz input. Blue is output.

Almach

That worked well, my images have all shrunk !

ICBM

Almach said:

Input at max output



That's around 140W ( handbook says 40W ! )

That's interesting. This is one of these "defies the laws of physics" amps as I'm sure you'd spotted from the stated power draw. I assumed that it was the output power that was falsely claimed to be 100W, not the input power being under-stated - since I would have thought that the input power claim *must* be correct to comply with regulations.

This is the second one I now know of which is definitely the other way round though - my Ampeg Micro VR, claimed output 200W, claimed input 150W, actually puts out 200W and draws 288W.

So if they're all claiming far less power draw than is actually the case, what happens when you've got a load of these things all running off one lightweight extension cable and the knowledgable user adds up the claimed power draws to work out if it's OK, and thinks it is...

DJH83004

Really interesting piece of work so well done for taking the time, What is unusual  is the the 192kHz sinewave present on the output. The frequency is typical of a switch mode power supply, but generally the chopper oscillators produce a square wave, which may get rounded through filtering, but not normally sinusoidal  - interesting. Regarding the power consumption labelling, the EU machinery directives for power consumption, used to be quite ambiguous and self policing (peak, average, percentiles etc.), so manufacturers would always err on the low side from an energy efficiency viewpoint (they were essentially written for washing machines and vacuum cleaners !!) I'm sure the guys on the forum who manufacture amps have a more up to date take on this than me,       

exocet

192khz looks suspiciously like a multiple of the sampling rate used? 

Actually forget that, just looked it up - it samples at 44.1 k although 192Khz sampling is used in audio industry so may be it's an artifact from the chipset used. Don't think you'd want that present in the output stage though.

monquixote

I shouldn't think the 192k thing is an issue.

I doubt a guitar speaker could actually produce it and even if it could it's massively higher than even what a dog can hear.

exocet

Hearing it isn't the issue. It's being amplified by PA stage so drawing power..at least that's my take on it. Actually I just re read the OP...the 192 k wave is only 300mv p2p so I agree it's not an issue. For some reason when I originally read it, I thought it said 192 k sine at cloise to full rated power.... like an RF transmitter!

Almach

Looks like the 192Khz is the carrier used by the PWM in the class D output stage. The web suggests available bandwidth is around 10% of the PWM chopper freq.

DJH83004

Ah, class D, didn't realise, as never worked on one, and to be honest never will as to me they are un-repairable, with all the SMT and my poor eyes and too many blown power mosfets! but does make a lot of sense.
  

ICBM

Although a frequency that high can't be reproduced by a guitar speaker, I wonder if it has any effect on how it works? I'm thinking something like the bias frequency in a magnetic tape recorder - If the speaker is being subjected to a small constant frequency it can't follow, does it improve how it responds to those it can? I have no idea...

Cirrus

ICBM said:

Although a frequency that high can't be reproduced by a guitar speaker, I wonder if it has any effect on how it works? I'm thinking something like the bias frequency in a magnetic tape recorder - If the speaker is being subjected to a small constant frequency it can't follow, does it improve how it responds to those it can? I have no idea...

I'd be surprised. Isn't the Bias frequency in tape a; loud in comparison to the signal and b; a way of aligning the random polarisation of the magnetic particles so they respond in a more uniform way to the signal you're recording?

With a speaker, the physical inertia of the cone is going to mean that a high enough frequency isn't going to be able to get it moving, I suppose if it's strong enough it is however going to be generating heat in the coil and increase the chance of it melting...