EEVblog Electronics Community Forum

Products => Test Equipment => Topic started by: rcbuck on November 09, 2017, 10:59:14 pm

Title: DSA815 Tracking Generator Damaged
Post by: rcbuck on November 09, 2017, 10:59:14 pm
Somehow I must have damaged the tracking generator output port on my DSA815. Attached are screen shots to show what I am seeing.

If I feed the tracking generator into the SA input, span set to 6-9 MHz, I see the following levels at these frequencies:

TG Level set to -10 dBm: 6.193333 MHz:  -8.86 dBm   6.206666 MHz: -9.86 dBm
TG Level set to -20 dBm: 7.293333 MHz: -18.4 dBm  7.306666 MHz: -19.4 dBm

If the TG Level is set to 0 dBm, the output level is flat across the entire 6-9 MHz range. If I set the start frequency to 5 MHz and stop frequency to 30 MHz, the output level is flat across the entire range. If I set the start frequency to 1 MHz and stop frequency to 5.5 MHz, I see a lot of level breakup at about 4.48 MHz.

The change when set to -10 dBm or -20 dBm is an abrupt change. It looks like this on the screen:

6.193333 MHz
------------------------
                              |  6.206666 MHz
                              -------------------------

If I set the start frequency to 5 MHz and stop frequency to 30 MHz, at 6.208333 MHz I see -9.03 dBm. If I turn the Marker knob one more click, the frequency changes to 6.250000 MHz and the level drops to -10.03 dBm. RBW and VBW are set to 300 KHz for all of these tests which is why the frequency numbers look strange.

I have a Tektronix AFG3021B. If I feed that into the DSA815 and set it to sweep from 6 to 9 MHz, I do not see the 1 dBm drop at any point in the scan. Also, a sweep from 1-5.5 MHz does not show any abnormalities.

I had 4.3 volts applied to the TG output jack for about 10 minutes. I was testing the gain of an amplifier and didn't realize the input and output switching relays have wetting current flowing through them. The TG output is labeled Max 50 V DC so it should have been OK. However, I am fairly sure that is when I started seeing the 1 dBm drop.

I could open the case and remove the RF shield enclosure and check for signs of damage. However, the warranty seal is still in place even though the warranty ran out years ago. I have contacted Rigol about a return for repair but haven't heard from them in the last 2 days. Apparently their service support isn't that good in the U.S. My unit is too old to upgrade for 10 Hz resolution so I am stuck with 100 Hz which is good enough for my uses.

I watched Dave's teardown and it looks like there is no capacitor protecting the output of the TG. It appears there is either a resistor or a diode connected from the output terminal to ground.

Has anyone else experienced a similar problem? If so, were you able to fix it?
Title: Re: DSA815 Tracking Generator Damaged
Post by: Orange on November 10, 2017, 03:38:23 pm
Looks OK to me; no worries. See below my DSA815 TG.
The step that you see are due to the calibration table of the TG
Title: Re: DSA815 Tracking Generator Damaged
Post by: rcbuck on November 10, 2017, 04:31:29 pm
Hmmm..., interesting. Does that mean the unit isn't really accurate below about 4.5 mHz? Do you see the same problem in the 6.19-7.2 MHz range that I am showing with screen shots 2 and 3? I am using the unit to test LC filters and expected to see a straight line across the filter's passband.

What or where is the calibration table you are referring to?

Thanks for your comments and screen shot.
Title: Re: DSA815 Tracking Generator Damaged
Post by: Orange on November 10, 2017, 05:37:29 pm
The system is accurate, but you need to normalize it first. This normalization, corrects the display for the errors in the output level of the TG

Connect the TG output to the input of the analyzer
If you press then TG button (TG on), choose Normalize, then normalize on, should give you a strait line.

All analyzers work like this.

The calibration table inside the DSA is set by the factory to correct the biggest errors in output level, but it does this with rather coarse steps of 1dB



Title: Re: DSA815 Tracking Generator Damaged
Post by: rcbuck on November 10, 2017, 08:37:18 pm
OK, thanks. I normalized it before running the sweep.

I heard back from Rigol Tech support. They said they were seeing the same thing on their unit. They sent a screen shot that looks similar to mine and yours. They also said the specs say the TG is flat +/- 3 dB. So I guess 1 dB easily meets their specs.

I wonder if the Siglent SSA3021X is any better?