A CLOSE LOOK TO LOW COST NOISE GENERATORS

General discussions
ON5HB
Posts: 140
Joined: Sat Dec 29, 2018 1:07 pm

Re: A CLOSE LOOK TO LOW COST NOISE GENERATORS

Post by ON5HB » Mon Feb 18, 2019 4:37 pm

To all.

Don't take life too serious.
If you get upset, count to 10 or walk the dog.

We are all here on the forum to help others.
If you do not like a post of somebody else, nothing stops you from ignoring it.

There is no point in getting upset over something you read on a computer-screen.

I'm not pointing at anyone.

Like the song: Don't worry, be happy!

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: A CLOSE LOOK TO LOW COST NOISE GENERATORS

Post by glovisol » Tue Feb 19, 2019 2:27 pm

Now that the dust has settled, I wish to post my sincere thanks to Roger, for bravely putting forward a different vewpoint, originating a discussion that, in perspective, I think made all the readers appreciate the value and the usefulness of the information today easily and freely available on Internet to us all. The perfunctory pinch of salt being always well advised.

This said, I wish to convey the following information, that I judge to be very important, even if it will appear obvious to people in the know. The ideal instrumentation for amplitude testing a passive filter is Tracking Generator + Spectrum Analyser, equipment costing upward of $1,500 /3,000, an expense out of proportion for any ham/hobbyst.

Now:

1) as far as the Spectrum Analyser is concerned, we have the RSP-1A and/or the RSP2 and/or the RSPduo PLUS the EXCELLENT RSP-SAS software being developed by Steve (whom I do not hesitate to call our Benefactor) which I have used in the Filter threads. A new and far more complete version of the RSP-SAS software should be available very soon, placing this set-up on par with much more expensive equipment.

2) as far as the Tracking generator is concerned, Steve and others are at work to produce a workable unit using an existing synthesiser board. But what can be used as a substitute in the meantime? The obvious reply: a wideband Noise Generator.

3) After the discussion with Roger et al, I have spent the last three days roaming over the Internet to find a Noise Generator which could replace the presently discussed BG7TBL. I can only conclude what I already knew: THERE IS NO REPLACEMENT AVAILABLE because all that is available is either too expensive, $600 and up, OR does not have a power output even remotely approaching that of the BG7TBL, e.g. -40 to - 50 dBm, which is ideal for passive filter testing. See the second post of:

https://www.sdrplay.com/community/viewt ... f=5&t=3352

So the conclusion is: if you need to sweep passive filters and you do not wish to sacrifice an arm and a leg, for the time being use an $15 Noise Generator BG7TBL, fed with a supply voltage of 9 V. If you need a precise wideband assesment of amplitude, first make a screen photo of the generator's output connected to the RSP-1A thru a suitable attenuator. Then do your measurement and apply a suitable correction.

I really hope someone will come along, explain that I am wrong and SUGGEST A MORE SUITABLE REPLACEMENT even costing $100 to 150 ( I wish to ruin myself!). Nothing would make me happier!

glovisol

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: A CLOSE LOOK TO LOW COST NOISE GENERATORS

Post by glovisol » Thu Feb 21, 2019 2:44 pm

OUTBOARD OUTPUT AMPLITUDE EQUALISER FOR THE BG7TBL NOISE GENERATOR 3 TO 35 MHZ

It is possible to remove the main problem (output levelling) of the BG7TBL, limited to the frequency range that interested me for filter development, e.g. 3 to 35 MHz, with a simple outboard L/C network. In reality equalisation should work down to 2 MHz, but it is difficult to say if the drop is due to the generator or the measuring software down there. I have measured the equalised output of the repaired noise generator. e.g. the one with the last amplifier removed and with a jumper from the second amplifier to the output SMA. Output level shown in the pics is 6/7 dB less because of the averaging action(level 16) of the display.
Attachments
EQUALISER 1.jpg
EQUALISER 1.jpg (193.33 KiB) Viewed 3873 times
EQUALISER 2.jpg
EQUALISER 2.jpg (193.92 KiB) Viewed 3873 times
EQUALISER 3.JPG
EQUALISER 3.JPG (162.24 KiB) Viewed 3873 times

Reason: No reason

NeilP
Posts: 53
Joined: Wed Apr 04, 2018 3:11 pm
Location: Jersey

Re: A CLOSE LOOK TO LOW COST NOISE GENERATORS

Post by NeilP » Wed Mar 20, 2019 4:05 pm

I've got a noise generator you can have. she will even pay here own way...the wittering just :lol: never stops

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: A CLOSE LOOK TO LOW COST NOISE GENERATORS

Post by glovisol » Wed Mar 20, 2019 5:16 pm

Sorry for that....I guess there is no remedy. Much easier with the electronic counterparts....

Reason: No reason

someYguy
Posts: 128
Joined: Sun Apr 10, 2016 7:12 pm

Re: A CLOSE LOOK TO LOW COST NOISE GENERATORS

Post by someYguy » Sun Mar 24, 2019 4:13 am

Look here:

viewtopic.php?f=7&t=1562

Everything in that post was done with a BG7TBL running at 8vdc and it worked fine. Still does, in fact.

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: A CLOSE LOOK TO LOW COST NOISE GENERATORS

Post by glovisol » Sun Mar 24, 2019 5:47 am

Hi someYguy,

I was and am aware of your thread. In fact it is your thread which, on June 20, 2018, started me to write :

"Testing RSPduo receiver sensitivity with noise generator", 17434 views to date.

https://www.sdrplay.com/community/viewt ... f=5&t=3352

In the first sentence there I was just praising your posts on the use of the BG7TBL noise generator.

Therefore I was somewhat surprised when I saw all that criticism poured over the BG7TBL, not considering that it was capable of good service at an extremely low cost, as soon as one understood it had to be operated at a lower than published voltage.

I have used it lately for filter development and I am still using two specimens of it!

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: A CLOSE LOOK TO LOW COST NOISE GENERATORS

Post by glovisol » Mon Mar 25, 2019 2:58 pm

THE NEW SPECTRUM ANALYSER ALPHA SOFTWARE & NOISE GENERATOR DATA

VERY MANY THANKS TO STEVE!!

What better time to obtain Noise Generator accurate reference data using the new issue of Steve's wonderful software?

The following data has been obtained using a BRAND NEW BG7TBL Noise Generator fed by a stable 11 V supply.

Supply voltage: 10.96 V
Suppy current: 196 mA
Power output @ 50 MHz: -45 dBm

Uploaded below output between 2 & 150 MHz.

FIGURE 1 has been drawn with the Noise Generator directly connected to the RSP-1A input to obtain the absolute level of output noise.

FIGURES 2 & 3 have been obtained with a 43 dB attenuator in between.

Curves 2 & 3 can be used to calculate a correction factor formula vs. frequency and/or a frequency dependent amplitude equaliser.
Attachments
ABS LVL.jpg
Figure 1. Output level to RSP-1A with no attenuator to measure absolute output level
ABS LVL.jpg (101.74 KiB) Viewed 3363 times
2- 100 MHz.jpg
Figure 2. Output variation 2 to 100 MHz
2- 100 MHz.jpg (246.4 KiB) Viewed 3363 times
50-150 MHz.jpg
Figure 3. Output variation 50 to 150 MHz
50-150 MHz.jpg (243.21 KiB) Viewed 3363 times

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: A CLOSE LOOK TO LOW COST NOISE GENERATORS

Post by glovisol » Mon Mar 25, 2019 5:08 pm

THE NEW SPECTRUM ANALYSER ALPHA SOFTWARE & NOISE GENERATOR DATA

Uploaded below is lower frequency Noise Generator output (43 db Attenuator between generator & RSP-1A).
Attachments
1-10 MHz.jpg
Figure 4. Output variation 1 to 10 MHz
1-10 MHz.jpg (114.39 KiB) Viewed 3326 times

Reason: No reason

sdrom33
Posts: 128
Joined: Sun Dec 09, 2018 4:38 pm

Re: A CLOSE LOOK TO LOW COST NOISE GENERATORS

Post by sdrom33 » Thu Apr 11, 2019 7:25 pm

glovisol it seems to me you are using a switch supply with the noise gen. This makes a lot of bad hash. How can you do measurement?

Reason: No reason

Post Reply