DIYHiFi.org

For the sake of audio
It is currently Sun Dec 17, 2017 2:13 am

All times are UTC




Post new topic Reply to topic  [ 25 posts ]  Go to page 1, 2  Next
Author Message
PostPosted: Mon Apr 28, 2014 2:16 am 
Offline
Cow
User avatar

Joined: Sat Apr 22, 2006 3:02 am
Posts: 289
Location: Stuttgart & Saar
I recently bought an Agilent 89441A, that is a spectrum analyzer from DC to 2.6 GHz.
It consists of 2 units, a baseband unit that goes from DC to 10 MHz and a downconverter
for the rest of the range, which is probably of lesser interest to this audience.

The baseband unit is able to do fast FFTs over a 10 MHz span. Dynamic range is not
outrageous, but OK if you are not after -160dB harmonics.

It also has a signal source, so it can measure frequency response, has 2 receive channels
with cross correlation & such, can do wonders in GSM demodulation that nobody here cares
about, but everybody thinks it's of use only for cell phone companies, and these cell phone
companies are trying to get rid of that stuff to make place for their next gen. systems, LTE
or later, whatever.
That has a nice influence on prices.

The last weekends, I tried to pair my low noise preamplifier with that machine and today I
got the first results. The preamp averages 20 AD4898 opamps, each with 0.9 nV/sqrt Hz,
so we end up at 200 pV/sqrt Hz from close to DC to 1 MHz.

I designed it before I had the 89441A and I aimed only for 100 KHz BW, but with the 10 MHz
BW of the 89441A, anything less than 1 MHz would be simply inadequate. The design cannot
be pushed further, and with > 20 MHz sample rate there are probably no serious alias
problems to be expected.

(I had to remove all default low pass caps and the capacitors parallel to the feedback
resistors to make it flat to 1 MHz, but there is still a dB missing at 1 MHZ...)
The ADA4898 is _really_ unity gain stable. I love it.

I decided to measure the noise voltage of some batteries, what I had already done
with a different spectrum analyzer.
First example is 4 Sanyo Eneloops AA in series.
Brown line is thermal voltage over 60 Ohms, or abt. 1 nV/ sqrt(Hz)-
Blueish line is Amplifier input shorted, abt. 200 pV / sqrt(Hz)
Green line is the 4* Eneloop.

Walls built a system with cross correlation to -205 dBV, and got even better results,
but that would take another preamp. Probably I'll do it.


You do not have the required permissions to view the files attached to this post.

_________________
Everything has been said already - but not yet by everyone. (Karl Valentin)


Last edited by gerhard on Mon Apr 28, 2014 3:10 am, edited 3 times in total.

Top
 Profile  
Reply with quote  
PostPosted: Mon Apr 28, 2014 2:45 am 
Offline
Cow
User avatar

Joined: Sat Apr 22, 2006 3:02 am
Posts: 289
Location: Stuttgart & Saar
Since I can post only one attachment:

Tadiran Lithium battery, digi-key 439-1009-nd , mfgpn tl-5955/t, 3.6V, 2/3 AA size

I bought it as a replacement for my Hp noise factor thingie, It came in a
separate box from D-K, seems to be impossible to transport it on a passenger
plane ( wonder if they can transport hp8970Bs on a plane).

This one has a nice 1/f characteristic, probably constructed to deliver
10 uA forever, not too much at a time.

But remember, the brown line is about the performance of an AD797 or LT1028!

Seriously time to go to bed now, nearly 5 am locally...

regards, Gerhard


You do not have the required permissions to view the files attached to this post.

_________________
Everything has been said already - but not yet by everyone. (Karl Valentin)


Top
 Profile  
Reply with quote  
PostPosted: Mon Apr 28, 2014 9:15 am 
Offline
Dog
User avatar

Joined: Tue Dec 14, 2004 12:26 pm
Posts: 4981
Location: Lisbon, Portugal
Oooh, this is great.
Do you have an SLA battery around?
BTW, thanks for taking the time to post this.

_________________
Carlos Filipe

"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius - and a lot of courage - to move in the opposite direction." Albert Einstein


Top
 Profile  
Reply with quote  
PostPosted: Mon Apr 28, 2014 10:35 am 
Offline
Cow
User avatar

Joined: Sat Apr 22, 2006 3:02 am
Posts: 289
Location: Stuttgart & Saar
yes, but it's sulfated :-(
Maybe I'll check the battery of my motorcycle. I don't want to
buy a new one just for the measurement and then forget about
it again and let it die from self discharge.

I still have to do a lot of things to solve: FFT from dc to 100 Hz or so,
that changes the resolution -> less power per frequency bin -> traces
are off screen,

The analyzer has no logarithmic sweep-> setting it up under PC control
and fetching data per ftp, mount several FFTs together to get one picture etc.
Maybe cross correlation, the batteries seem to be close to even this
noise floor.

Shielding is also a problem. The preamp and the test objects are in
thick Hammond alu boxes, but just connecting the test object with
a semi rigid SMA cable bring a lot of spurs. I had to make a
second box wth BNC feed throughs for everything together
(but the analyzer) to get rid of most of them. I still don't know where
the 190 KHz come from.

I also plan to measure references and power supplies, also lab supplies,
have already quite a collection of candidates.

A lot to do.

regards, Gerhard


You do not have the required permissions to view the files attached to this post.

_________________
Everything has been said already - but not yet by everyone. (Karl Valentin)


Top
 Profile  
Reply with quote  
PostPosted: Mon Apr 28, 2014 3:22 pm 
Offline
Goat

Joined: Mon Nov 28, 2005 1:10 pm
Posts: 545
Thanks Gerhard for the insight.
So the eneloops unloaded are as low noise as it gets. But what about dynamic performance, say eneloops powering the analog supplies of a dac?


Top
 Profile  
Reply with quote  
PostPosted: Wed Apr 30, 2014 8:16 am 
Offline
Cow
User avatar

Joined: Sat Apr 22, 2006 3:02 am
Posts: 289
Location: Stuttgart & Saar
alexandre wrote:
So the eneloops unloaded are as low noise as it gets. But what about dynamic performance, say eneloops powering the analog supplies of a dac?


Loaded? Maybee this extended weekend.
I think their performance simply comes from the small source resistance.
F.Walls wrote that small currents and even charging would not change much, IIRC.
Google for 1133.pdf F.Walls chemical batteries time freq nist.gov, I don't have
it here.

Gerhard

_________________
Everything has been said already - but not yet by everyone. (Karl Valentin)


Top
 Profile  
Reply with quote  
PostPosted: Wed May 07, 2014 4:05 pm 
Offline
Cow

Joined: Tue Jan 15, 2008 12:09 am
Posts: 280
gerhard wrote:
I still don't know where the 190 KHz come from.

Turn off the lights in the house.


Top
 Profile  
Reply with quote  
PostPosted: Wed May 07, 2014 4:45 pm 
Offline
Cow
User avatar

Joined: Mon May 21, 2012 2:13 pm
Posts: 152
Eneloops are nice!
What is the frequency scale, is it logarithmic?
I wonder whether eneloops could be used as voltage references :) They keep their charge for awhile... Not that precise for measurement equipment, but quite low noise for the places which need silence...

_________________
I have no idea what i'm doing, and at this point i'm too afraid to ask.


Top
 Profile  
Reply with quote  
PostPosted: Wed May 07, 2014 7:04 pm 
Offline
Cow

Joined: Tue Jan 15, 2008 12:09 am
Posts: 280
For use strictly as reference: a Lithium primary battery (non-rechargeable) holds the charge more time than the Eneloop. And I think has less noise.


Top
 Profile  
Reply with quote  
PostPosted: Thu May 08, 2014 4:26 pm 
Offline
Cow
User avatar

Joined: Sat Apr 22, 2006 3:02 am
Posts: 289
Location: Stuttgart & Saar
SoNic wrote:
For use strictly as reference: a Lithium primary battery (non-rechargeable) holds the charge more time than the Eneloop. And I think has less noise.


The plots above make it quite clear that the Lithium battery has _much_more_ noise, not less.

_________________
Everything has been said already - but not yet by everyone. (Karl Valentin)


Top
 Profile  
Reply with quote  
PostPosted: Thu May 08, 2014 4:43 pm 
Offline
Cow
User avatar

Joined: Sat Apr 22, 2006 3:02 am
Posts: 289
Location: Stuttgart & Saar
s3tup wrote:
Eneloops are nice!
What is the frequency scale, is it logarithmic?
I wonder whether eneloops could be used as voltage references :) They keep their charge for awhile... Not that precise for measurement equipment, but quite low noise for the places which need silence...


You can tell that from the form of the green trace in the second picture :D
On a log scale, the 1/f noise would be a straight line, not so curved.
So, the answer is: linear.

It seems, the analyzer can display log scales only from the result of a single FFT.
I want 0.1 Hz to 1 MHz in one picture, that would require one HUGE FFT.
Best I got upto now is 1.3 KHz to 1MHz in log, the effects of displayed points,
time series length, bandwidth, sample rate etc look somewhat complicated.
Probably I'll have to do several FFTs with different sample rates and combine
the results. Might be a lot of programming.

Walls found that even slight charging does not spoil the noise of Ni-Cadmiums, IIRC
So, that might be possible.

regards, Gerhard

_________________
Everything has been said already - but not yet by everyone. (Karl Valentin)


Top
 Profile  
Reply with quote  
PostPosted: Thu May 08, 2014 4:56 pm 
Offline
Cow
User avatar

Joined: Sat Apr 22, 2006 3:02 am
Posts: 289
Location: Stuttgart & Saar
SoNic wrote:
gerhard wrote:
I still don't know where the 190 KHz come from.

Turn off the lights in the house.


I did that already. Maybe it's one of the rigs on the workbench.
The energy saving lamp on the bench produces a carrier that
sweeps between 50 and 60 KHz. Quite fast. Impossible
to recognize on an analog spectrum analyzer.

I have a 3 GHz pulse generator from Anritsu that produces
ps-risetimes and ultraflat level, no idea how they can do that.
But to power it, I need the cable of the electrical loan mover
to plug it in - in the kittchen or in the cellar below.
Its power supply is sooo dirty!

Maybe I should buy an isolation transformer and some filters.
Would have to be big.

_________________
Everything has been said already - but not yet by everyone. (Karl Valentin)


Top
 Profile  
Reply with quote  
PostPosted: Thu May 08, 2014 6:37 pm 
Offline
Dog
User avatar

Joined: Tue Dec 14, 2004 12:26 pm
Posts: 4981
Location: Lisbon, Portugal
gerhard wrote:
Walls found that even slight charging does not spoil the noise of Ni-Cadmiums, IIRC


Batteries behave as big caps, they attenuate the noise of the charger.

_________________
Carlos Filipe

"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius - and a lot of courage - to move in the opposite direction." Albert Einstein


Top
 Profile  
Reply with quote  
PostPosted: Fri May 09, 2014 2:49 pm 
Offline
Cow

Joined: Tue Jan 15, 2008 12:09 am
Posts: 280
gerhard wrote:
Maybe I should buy an isolation transformer and some filters.
Would have to be big.

I would think that filters are implemented inside of those performance measuring tools.
Sure I would add a filter to your rig if that equipment is your bread maker.

BTW: When I was working in a national metrology lab, our best measuring tools had rechargeable sources (Ni-Cd or Lead-Ca) and we did unplug them from mains while in use.


Top
 Profile  
Reply with quote  
PostPosted: Fri May 09, 2014 3:37 pm 
Offline
Dog
User avatar

Joined: Tue Dec 14, 2004 12:26 pm
Posts: 4981
Location: Lisbon, Portugal
SoNic, please don't mess this thread up.
Go see if your batteries are charged and if it's raining outside.
Thank you in advance.

_________________
Carlos Filipe

"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius - and a lot of courage - to move in the opposite direction." Albert Einstein


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 25 posts ]  Go to page 1, 2  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group