DIY Audio Analyzer with AK5397/AK5394A and AK4490

Administrator
Joined 2004
Paid Member
Yes, it is still a very good deal. For me it was a question of money only. Miss the deal and never get one. I got one. I've used it nearly every day since it arrived.

The companion software, M.I., integrates with the RTX and controls it. It even has a scheduler to automate commands so it will run a series of tests with one command. Another very, very good deal even at list. If at all possible, get them both. You need not buy them at the same time.

-Chris
 
XFR spurious coupling issue?

Fitted screening cans (but RC mod proved beyond my tools so couldn't do that unfortunately). Then did some basic tests to ensure kit OK. During that I noticed several high harmonics of the main freq (50Hz for me) when set on most sensitive input setting with inputs all shorted at the input connectors (running AudioTester) Then did dome basic 'harmonic' hunting'. Lifting out the mains XFR and orienting it vertically with the back wall helped quite a lot but hanging it outside the side of the case (as far as the cables will allow) made a major difference to the extent that the harmonics are all but invisible. All rather indicative of spurious flux coupling from the XFR but will investigate further. Anybody seeing anything similar? Not sure what my next move should be.
 
Actually I see that 'xfe' and others are already discussing the mains XFR and possible spurious flux coupling matter. For what its worth I would like to stay with the 'old fashion' mains toroid solution even if it means moving it externally (as my previous post strongly suggests would be very beneficial). I'm not anti SMPS in fact I love them. I used to design them for aerospace use, light weight, efficient and can be made very low noise (if you know what your doing). However difficult to design and certainly non trivial to get safety etc approved for mains use.
 
XFR spurious coupling issue?

If I have got the hang of how to upload attachments this should show 3 pics of the mains XFR as factory fitted,vertically up against back panel and hanging outside of side panel. Significant differences.
 

Attachments

  • rtx various XFR locations.pdf
    960.3 KB · Views: 154
Member
Joined 2004
Paid Member
For magnetic induced hum fields shorting the input would be worst case since the current flows without limit. I mentioned earlier we need to find a steel (or mumetal) can we can put the transformer in. I think its too big for a tuna can and those aren't steel any more I think. . . Do you have the dimensions of the transformer? There are can's on eBay that might work.
 
XFR spurious coupling issue?

I have already tried a pretty thick(gauge) umetal 'collar' enclosing the XFR and protruding above the XFR height with XFR in its factory location as an 'easy fix'. Its not a fully enclosed can of course but thought it would be indicative of the way to go but it didn't have much effect hence then experimenting with the XFR location.

I'm of the school that says careful design of magnetics and their locations is very important (I have designed high power magnetics in aerospace SMPSs) but the best and cheapest / free way to get significant improvements is distance.
 
Member
Joined 2009
Paid Member
The diameter of the transformer is specified as 72 mm. The height is just over 40 mm.
I can see that protective covers with dimensions of 80x40 mm with an iron thickness of 1 mm are available. They might be suitable for testing. I have no idea about the shielding effect though.
Please remember: do not short circuit the transformer with the shield and mounting!

I have ordered a small number of transformers with magnetic shielding (GOSS band) and lower flux.
My intention is to compare the magnetic fields from this transformer and the original transformer. The diameter will increase to around 75 mm and the height will increase as well.
 
JensH. Yes a good reminder to be sure any experiments do not result in a shorted turn.

Out of interest I have just tried sitting the XFR in a steel tuna can with XFR in its factory location. Top is open but as an experiment it showed little improvement. An admittedly poor fit lid fashioned from mu metal (so XFR was crudely fully enclosed with gap for lead outs) also didn't help materially. Tuna can thickness isn't much but then a previous experiment with a reasonably think mu metal collar around the XFR didn't help much either. Intriguing.
 
The diameter of the transformer is specified as 72 mm. The height is just over 40 mm.
I can see that protective covers with dimensions of 80x40 mm with an iron thickness of 1 mm are available. They might be suitable for testing. I have no idea about the shielding effect though.
Please remember: do not short circuit the transformer with the shield and mounting!

I have ordered a small number of transformers with magnetic shielding (GOSS band) and lower flux.
My intention is to compare the magnetic fields from this transformer and the original transformer. The diameter will increase to around 75 mm and the height will increase as well.


Interested in how this turns out. If you find a good result, please let us know where we can get the transformer.


Cheers
 
Member
Joined 2009
Paid Member
If it is done correctly and the safety aspects are handled correctly, I don't see any technical problems in doing it. Make sure that it is done in a safe way to avoid any safety hazards. And use the correct wire gauge with stable connections.
You will of course void the warranty if still in force.
 
As JensH says putting XFR externally (in a steel box ideally) is straight forward but you must get the safety aspects correct for the external box and you also need to consider how to ground the RTX as the ground connection wont be into the RTX via the IEC connector. Can be done as a temporary fix such that the factory config is readily reinstated if required.