ASR dummy load configuration

[March Audio]

Again, the resistance in a dummy load doesn't matter. We don't care of the load is 4 ohm or 4.01 ohm. What we care about is whether there is a characteristic that changes with voltage. Maybe electronic welding is good. Maybe it is not.

Also, I connect and disconnect these cables hundreds of times. I routinely break cables and have to buy new, or make new. With a welded connection I can't fix it if it breaks.

So for everyday use they may be excellent. Testing though is a different matter.

As alluded to above this does present an interesting question as to how amps behave with a real speaker load and its VCR. Obviously all speakers will be different, but it might be an interesting experiment looking at the distortion of various amps with a test speaker compared to the dummy load.

Unfortunately I see no way of separating this VCR distortion effect from the distortion effects that any complex reactive load (speaker) effect may be having.

 

[amirm]

Real speakers likely have good bit of VCR effect. I will see if I can test it at some point.

 

[restorer-john]

They are ultrasonically welded.

Tell me more about that process. They look compressed and almost 'spot welded'.

 

[restorer-john]

Real speakers likely have good bit of VCR effect. I will see if I can test it at some point.

Dynamic compression plus overall impedance changes.

I know plenty of my smaller two ways exhibit this phenomenon once I heat up the voice coils a little. Take an impedance sweep on a hot (not burnt out!) speaker vs a cool one.

 

[March Audio]

Tell me more about that process. They look compressed and almost 'spot welded'.

Pretty much, via ultrasonic vibration.

https://www.bluejeanscable.com/articles/ultrasonic-welding.htm

https://en.m.wikipedia.org/wiki/Ultrasonic_welding

 

[Murrayp]

May I ask whether adding some load reactance and back emf mightn't give quite different and likely much poorer measured distortion results? Ideally of course the perfect amplifier will handle this, but one might expect a semiconductor device trying e.g. to drive a positive voltage when the current flows in the opposite direction, must present an interestingly different gain for the control loop to mange? Once I tested the square wave response of a zero feedback mosfet follower valve hybrid amplifier into resistors - the response was beautiful, with rather tidy and fast rise and fall times. When I connected that same amplifier output to my speakers rather than resistors, the waveform was no longer recognisable as a square wave! Nevertheless the amplifier sounded quite nice to me, if a bit "soft". Are we maybe letting amplifiers off far too easily by not incorporating real load reactances and back emfs? I can't imagine this being a lesser effect that those created by imperfect connections? Again - just my 2c - I'm here to learn - thanks!

 

[March Audio]

You might be interested in this thread
https://www.audiosciencereview.com/...g-a-reactive-load-for-amplifier-testing.7934/

 

[FrantzM]

This kind of enlightening discussions is leading me to bump up my monthly donation to ASR.

Damn you People!!!

 

[Murrayp]

You might be interested in this thread
https://www.audiosciencereview.com/...g-a-reactive-load-for-amplifier-testing.7934/

oops - excuse my inattention - quite right

 

[LTig]

[..]
As it turns out, I had used very high quality/precision Dale resistors in my headphone amplifier load. I thought at the time it was overkill as one does not need precision there but I spent the money on them anyway. That likely enabled me to measure headphone amplifiers with such low distortions.

As I understand this problem is related to the current. Would it be sufficient for a (headphone) load to just use 10 or 20 standard resistors (1% or maybe 0.1%) in parallel to split the current load and prevent the distortion? I can get 0.1% resistors for 20 or 30 €-cents ...

 

[amirm]

The issue I discovered for speaker testing was voltage dependent, not current. Although current indirectly contributes probably.

Turns out for headphone testing voltages are low enough that VCR is not a factor. My latest high-precision load uses random resistors but with Kelvin connection and no switches, jacks, etc. This outperformed my 1% resistors with switches and such in between.

Ultimately though, you have to build these things and test them as there is just no data on it.

But yes, I have seen people build even speaker load out of a large series of lower wattage resistors. And I was thinking about along the same lines as you if one was to make a production out of it.

 

[LTig]

The issue I discovered for speaker testing was voltage dependent, not current. Although current indirectly contributes probably.

Yes, of course, it's called VCR and not CCR - have to read more carefully.

Turns out for headphone testing voltages are low enough that VCR is not a factor.

Well then, would using many resistors in serious also fix the problem because each resistor sees only a small voltage?

My latest high-precision load uses random resistors but with Kelvin connection and no switches, jacks, etc. This outperformed my 1% resistors with switches and such in between.

Ultimately though, you have to build these things and test them as there is just no data on it.

How can I separate THD induced by VCR from THD created by the DUT?

I could connect a 1:1 voltage divider to the output of my RME, connect one of its inputs to this output and the other input to the voltage divider. Then run a loop back test and compare the THD readings of both channels.

 

[amirm]

How can I separate THD induced by VCR from THD created by the DUT?

By knowing that with another load, it does a lot better. That is how I discovered the issue. So you really need a known, low distortion device to experiment.

 

[LTig]

I built a 1600W 8/4 ohm dummy load using 32 cheap nomame 100W resisors from ebay mounted on a big fan cooled heatsink.

I've ordered 16 1Ohm 50W resistors (ARCOL HS50) but I have difficulties to find affordable heatsinks (EU). Where did you get your headsink and how much did you have to pay for?

 

[pma]

I use this, 10ohm//6.8ohm makes 4.05 ohm. Heatsinks are available at Fischerelektronik, RScomponents and many other places. Take care about inductances.

 

[LTig]

I use this, 10ohm//6.8ohm makes 4.05 ohm.

How big is your heat sink? Can you feed it 300W continously, and if not for how long before it gets too hot?

Heatsinks are available at Fischerelektronik, RScomponents and many other places. Take care about inductances.

Do they sell to private customers? The times were I could order stuff via my employer are long gone - SAP/SoX makes it too expensive to sell to private people.

 

[pma]

For 300W continuous my heatsink would be small.

Check this

https://uk.rs-online.com/web/c/hvac...ort-by=Thermal Resistance&sort-order=asc&pn=1

 

[Armand]

I've ordered 16 1Ohm 50W resistors (ARCOL HS50) but I have difficulties to find affordable heatsinks (EU). Where did you get your headsink and how much did you have to pay for?

I used something I had salvaged from some old power electronics. I never throw anything away and have a garage attic to prove it
Using a high speed fan helps a lot. I use a temperature controlled fan on my load.
I have dumped 1200W power into it for 2 hours. The temperature reached about 100 degrees celcius.

 

[Armand]

What about RS?
This one has 0,13C/W for up to about 1kW of power.
https://no.rs-online.com/web/p/heatsinks/6920607

 

[LTig]

What about RS?
This one has 0,13C/W for up to about 1kW of power.
https://no.rs-online.com/web/p/heatsinks/6920607

That's probably too small to mount 16 resistors, and too expensive for 8 since then I need two of them.