Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error

Options
View
Go to last post Go to first unread
jmarvin  
#1 Posted : Sunday, March 8, 2020 6:47:46 AM(UTC)
jmarvin

Rank: Member

Groups: Member
Joined: 2/22/2020(UTC)
Posts: 3
United States
Location: Colorado

This is my first post in these forums. I recently bought the full Buffalo 3SE Pro 9038 kit, along with a Cronus and Hermes BBB interface. I just got the power supplies, Mercury, and the DAC card all set up, and did some initial testing.

I hooked up a SPDIF connection from a PC which was sending a full scale stereo sine wav, and things are basically working. However, I noticed when I raised the DAC volume level such that the peak voltage level crossed about 2.1V (full scale peak voltage when the DAC volume is all the way up is about 4,2V), all of a sudden a lot of harmonics poked up. I checked to make sure there was still some headroom on the Placid (i.e. shunt level hadn't gone down to far), but since this isn't a dynamically changing source, it doesn't really need a lot of headroom.

Anyway, I'm not really concerned right now, for two reasons:

1) My measurement methodology is highly suspect. Although the source signal looks really clean, I'm using a Digilent Analog Discovery 2 USB scope, which is certainly not the best for high resolution measurements (it's got a 14 bit A/D converter). It's quite possible that corssing the 2.1V line causes an issue in the USB scope, and nothing is changing in the Mercury output. I've also got a bunch on temporary, potentially suspect, wiring, and am using the SE outputs. On the other hand, nothing else is changing at that point, i.e. I have a clean signal at 2V peak, and it gets harmonically messy at 2.1V peak by just changing the DAC volume.

2) Even if the problem is real, and I can't find an easy fix, I can probably live with 2V output. I'm pretty sure that will drive my system to high enough listening levels.

In the future I will make better measurements, probably using a higher end sound interface, like my Focusrite 18i20 which can easily handle the peak output from Mercury (not sure I have any other easy ways to do this, since I don't have easy access to high quality test hardware), i.e. just take a long stereo sample at 96/24 and then run a static spectrum conversion with known good DSP software.

However, I wanted to post this just to see if there is something I should check, or if this isn't a surprise, or something else I am missing.

jmarvin  
#2 Posted : Monday, March 9, 2020 10:37:52 AM(UTC)
jmarvin

Rank: Member

Groups: Member
Joined: 2/22/2020(UTC)
Posts: 3
United States
Location: Colorado

I did measure the balanced outputs of the mercury and this issue doesn't appear to happen there. It only appears to happen on the SE outputs. I haven't seen a schematic for the Mercury. If it is anything like the Ivy schematic, the LM47920 op amp that converts the balanced outputs might be the only thing between the balanced outputs and the SE output, other than a resistor. If so, is there any chance the LM47920's input is being overloaded? It has a max rating of 0.7V on the input. The SE output is putting out 2.1V RMS when the harmonic distortion occurs, while the balanced outputs are putting out half that on each each output (1.05V RMS). The inputs to the LM47920 are +/- 525mV, so it looks like the op amp is configured with a gain of 2 (I think the IVY is the same). As far as output voltage goes, everything is fine, but I can't find anything in the spec regarding distortion based on input voltage as it approaches the max input.

The max spec is 0.7V, but that is the potential damage point. One place in the spec listed a typical input of 0.1V. I don't know if 0.525V may be a problem. Note that this is not a small amount of distortion. It is primarily odd harmonics all the way up the spectrum, and the 3rd harmonic is only about 27 dbV down from the primary signal. The distortion goes down to acceptable levels almost on a knife edge when lowering the volume a tiny bit from this point, i.e. overloading the input of the LM47920 might be reasonable cause of this.

Note that I am certainly not an expert in analog circuit design, and know just enough to be dangerous. :)

EDIT: It just occurred to me that 0.525V RMS is over 0.7V peak.

Edited by user Monday, March 9, 2020 4:56:14 PM(UTC)  | Reason: later observation

Russ White  
#3 Posted : Tuesday, March 24, 2020 7:18:32 PM(UTC)
Russ White

Rank: Administration

Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN

Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
The op amps are all being used in spec. The voltage difference you speak of will never exceed more than .1V in the worst case at the opamp input pins :) When I have measured the distortion and it is very nearly identical to the balanced output.

I think you can safely relax and enjoy the music :)

Cheers!
Russ
Rss Feed  Atom Feed
Users browsing this topic
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.