Overcomplicating a dummyload…

One of the things I’ve never gotten round to building is a headphone amp dummyload. And to be honest it is also a thing that I haven’t been missing much – at least until I got an oscilloscope…

Usually a dummyload is just a pair of resistors (hence the name…) and I have seen a few designs previously that were just “resistors in a box” (most notably this old one from Tangent which was probably the first one I ever saw). There are also other versions (and even some commercial ones) but a couple of months ago my gears started turning that maybe there was a better way to do this (probably while looking at this one actually).

Part of what I wanted to address in my version is the number different impedances that were required to perform good and exhaustive tests. Low-impedance phones is pretty much the norm these days, but Sennheiser had 300 ohms for their higher-end models for a long time and I still (occasionally) use my BeyerDynamic T1s as well (which are 600 ohms). At the other end of the scale there is also some movement because when I started out in headphones a pair of 32 ohms Grados were considered “low impedance” but now you regularly see even high-end headphones going lower than that (down to around 16 ohms).

A quick jot-down of “standard” values turned out to be 5-6 obvious ones and then using a standard 6-position rotary switch as the “centerpiece” was an obvious decision. I originally envisaged this as a stereo board with axial resistors and after researching parts some more I found a series of 3W/1% tolerance resistors that seemed perfect for the job – and incredibly they come in precisely the values I wanted, so the only deviation is the resistor tolerance. Higher-power resistors probably wouldn’t have been amiss, but on the other hand 3W is a lot for most headphone amps and I am pretty sure that the resistors will take a bit than that for short periods if needed.

I’ve also added a couple of positions for (switchable) load capacitors to simulate cable capacitance, but obviously this is less standardised so I’ve just chosen a couple of values that I found reasonable. Last, but not least I added a “low-range” option with a resistor in parallel which extends the range downwards. With the values given this means 16-600 ohms + a 10k line-level setting. The resistor values are obviously freely selectable, but I think I’ve managed to find a range that works really well.

The downside of the stereo board is that there is no direct indication of which load is selected. That could be fixed by reducing the number to positions to 4 (and using a three-way switch), but I didn’t want to do that. Instead I opted to make a mono version of the board and used the second deck of the switch to select LEDs as well. Obviously that means the board has to be powered in some way, but it still seems reasonable to do. Originally I had individual resistors for each LED to be able to use different colors, but in the end I decided that was not necessary anyway (as long as you are smart about your choice of LEDs 😀 ).

I’ve just had a quick test of this and both versions seem to work really well. It’s definitely something I need to mount in a more permanent way for lab use. The missing part on both board versions in the pictues is the 10k resistor to simulate a line-level load. It’s not “missing” as such, but I managed to order a wrong part which doesn’t fit the board. Rather than making it fit I thought I would just wait a little get the right one with my next parts order 🙂

2 Responses to Overcomplicating a dummyload…

  1. Pingback: Project files: Headphone amp dummy-load… | theslowdiyer

  2. Pingback: Measuring the SlowMoy OPA1656 – (mis) Adventures in Hi-Fi

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: