Marti 950 MHz STL-10 spurious emissions

Gotta watch those Fresnel zones!
Post Reply
kcbooboo
Posts: 455
Joined: Sat Dec 10, 2011 4:53 pm
Location: East of the Mississippi

Marti 950 MHz STL-10 spurious emissions

Post by kcbooboo » Sun Nov 23, 2014 10:39 am

I've got a pair of Marti STL-10 / R10 systems here around 949 MHz. Making 10 watts into a good dummy load.

I connected the RF output to a 50w 30dB coaxial attenuator followed by a 2w 20dB coaxial attenuator, to the input of my spectrum analyzer. The SA claims the 948 MHz carrier is at -9.75dBm, which is close enough for me. 10w is +40dBm and I've got 50dB of attenuation so as to not overload the SA. -10.0dBm would be the target.

These units generate and amplify the signal at 1/2 the desired output frequency. They then run it through a varactor doubler to come up with the 950 MHz signals.

There's a sub-harmonic at 474 MHz that's at -66dBm, which is about -56dBc. The Marti specs call for -60dBc. I also checked up around 1422 MHz (3 times 474 MHz) and found a similar -66dBm spike. My SA is only good to 1.5 GHz so I can't check any higher.

Both units have the same levels at the output frequency and +/- 1/2 the output frequency. Either they're both tuned wrong or that's just how they are.

Is this normal? Suggestions to minimize the out-of-band signals besides some sort of bandpass filter? Does anyone have anything to compare this to?

Bob M.

DaleNX8J
Posts: 2
Joined: Sat Jan 13, 2018 5:56 pm

Re: Marti 950 MHz STL-10 spurious emissions

Post by DaleNX8J » Sun Jan 14, 2018 7:15 pm

kcbooboo,

Here it is 3 years after your post.... but this is still relevant.

If an XMTR does not make published spec, it's either your test equipment/setup, a problem with the XMTR, or specmanship. We use STL-10's for an AM link but I've never put them on an analyzer.

To allow folks to sleep easier, consider that the antenna likely shows loss at half-freq, so the sub at 474 MHz is likely radiating beneath allowable spec limit. Could be the same for the 2nd harmonic, if the antenna is selective enough.

If I had this unit on the bench, I'd look closely at tuning around the doubler and PA, while operating into a known 50 Ohm dummy. Lots of things can cause a dirty signal; poor connections, bad power supply bypass, RF feedback, loose screws on a shielded enclosure, etc.

User avatar
KPJL FM
Posts: 573
Joined: Fri Nov 16, 2007 9:28 am
Location: planet Earth, 3rd rock from sun

Re: Marti 950 MHz STL-10 spurious emissions

Post by KPJL FM » Mon Jan 15, 2018 9:22 am

In order to check any RF generator for spurs on a SA, you need to kill the Fo completely at the input to the SA. Otherwise, the SA will show harmonics up the wazoo.
Trim to fit, paint to match, tune for minimum smoke.

kcbooboo
Posts: 455
Joined: Sat Dec 10, 2011 4:53 pm
Location: East of the Mississippi

Re: Marti 950 MHz STL-10 spurious emissions

Post by kcbooboo » Tue Jan 16, 2018 6:41 am

I feel reasonably confident that with the 50dB of attenuation ahead of the SA, the harmonics I was seeing are truly present. When I increased the input signal by removing some attenuation, the fundamental and all harmonics increased by exactly the same amount. Usually SA overload or internally-generated spurs will not change by the same amount as the fundamental when the signal level is changed, especially by 10-20dB.

Bob M.

Post Reply