Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Behringer HD400 measured (Read 2784 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Behringer HD400 measured

I hadn't seen any measurements for these units posted anywhere (only hints that there was little to find fault with), so here you go.
Output: Audigy FX Front Headphone @ level 92 (-1.5 dB vs. 2.0 Vrms, so -3 dBFS sine @ -4.5 dB), 22 ohms
Input: Audigy FX Rear Line-In @level 3 (+2.3 dB nom), imp unsure (10k? 22k?)
Cables: 1/4" to 2x TS 3 m, 2x TS to 2x RCA 3 m + stereo RCA to 1/4" adapter
RMAA 6.3.0 @ ASIO via ASIO4All 2.10

I ran the tests in both directions:
"Forward" - FX output to HD400 input, HD400 output to FX input
Results "forward" (SAV)
"Reverse" - FX output to HD400 output, HD400 input to FX input (it's just a dumb 1:1 transformer, right?)
Results "reverse" (SAV)

Not reproduced here are the results for my first attempt, which involved using the rear line-out and front mic-in - let's just say an input impedance of 2.2 kOhms connected to +3.2 V made things look AWFUL. Distortion beyond belief, lousy bass response, in short: Running into saturation bigtime. So whatever you do, keep DC away from these things. You didn't expect super big cores at the price, did you?

Thankfully the results under normal operation conditions look a lot better.
For one thing, there is no sign of any of the typical ground loop garbage, regardless of the considerable length of cables (the low-level tones seen are always present). So it's doing its job as an isolation transformer just fine. (I have no idea how to easily quantify unwanted coupling across the transformer though. Measure CMRR?)
The swept sine test (SAV only) indicates the expected quickly-rising low-frequency distortion (dominant 3rd) which exceeds -60 dBr at 40 Hz. I can live with that.
There is little influence on midband distortion.
We've got some (presumably inductive) coupling between the channels that reaches -42 dB at 20 kHz and gets even worse in the ultrasonic region. Fine for stereo, possibly problematic if you want to run independent channels over it (or both an amplified and unamplified version of the same signal).
Curiously, we are seeing some differences between "forward" and "reverse" operation in the higher frequencies. "Forward" operation shows a very well-behaved response with very little peaking (about +0.1 dB) while "reverse" displays an ultrasonic peak - a deviation of +0.4 dB at 20 kHz should still be easily acceptable though. Ultrasonic crosstalk also gets worse in "reverse". In return, high-frequency distortion / IMD in "forward" mode is degraded a fair bit - which makes me think that the transformer must present a far lower-impedance load to the output there, making the MAX97220A break a bit of a sweat. (People have also reported low levels with high-Z sources like a guitar.) Any xfmr gurus around who could tell what that means?

In short, while this device has its limits, it is quite usable indeed. At the levels I run into my speakers (rarely higher than -10 dB re: 2.2 Vrms, minus 3 dB minus (-ReplayGain)), I see no reason to be concerned about either low-frequency or high frequency distortion. (My speakers are far worse anyway.) Same goes for xtalk in stereo operation. When running full consumer line level or even more across it, I'd probably wire it up in reverse though.

Note that the jack contacts are obviously not gold-plated and may give better wear characteristics with plain nickel-plated connectors (on the connectors, I mean - gold is fairly soft and easily wears). (Galvanic corrosion shouldn't be an issue with either nickel or gold.) The jacks also are merely soldered on and not anchored otherwise, so a bit of care when plugging things in is advised. (If you want something more robust, you need to shell out more money, simple as that.)
Also note that it's not a DI box and doesn't work well as such (passive DI boxes have turns ratios of more like 10:1 in order to provide a high-Z input).

Behringer HD400 measured

Reply #1
For reference, here's a direct connection (SAV) between front headphone out and rear line-in, using a mini to mini cable about 1.5 m in length.
There is some (though not extreme) ground loop noise present now. This allows a glimpse at the effectiveness of the HD400. A spot check with the tone at 23470 Hz shows a reduction from -108 to -126 dB (reverse operation is about 2.5 dB worse), while at 3 kHz we are looking at -133 dB vs. -113 dB (with only half a dB worth of difference between forward and reverse, though things are so close to the noise floor it's hard to tell).

I may need to find a way of creating (and then eliminating) a more severe ground loop.

Behringer HD400 measured

Reply #2
Its possible they put a 20 Khz low pass filter on the input side to reduce ultrasonic nonlinearity in the transformer.  That might explain why flipping the directions changes the frequency response a little.