DC Block response

[Jester]

I picked up one of these on eBay  http://www.ebay.com/itm/Attenuator-DC-Block-N-male-to-Female-DC-6-0GHZ-50ohms-RF-coaxial-Power-2W-200V-/361279079925?hash=item541de945f5:g:XAQAAOSwPgxVOetN

Initial test was with a signal generator and scope, it blocks DC as advertised

Next test I swept on the SA and it looks like this, what am I doing wrong?




 

[nctnico]

Your screendump doesn't make any sense. The centre frequency is supposed to be 15MHz but your marker shows 89kHz. Perhaps you should reset the spectrum analyser and test again.

[hendorog]

Your screendump doesn't make any sense. The centre frequency is supposed to be 15MHz but your marker shows 89kHz. Perhaps you should reset the spectrum analyser and test again.

It's a log frequency scale - the 'centre' value is 1.5 divisions in from the Right edge of the screen!

[nctnico]

Your screendump doesn't make any sense. The centre frequency is supposed to be 15MHz but your marker shows 89kHz. Perhaps you should reset the spectrum analyser and test again.

It's a log frequency scale - the 'centre' value is 1.5 divisions in from the Right edge of the screen!

I missed that! But appearantly so did to OP because it would mean the left part of the screen is in the Hz range which is where (AFAIK) all spectrum analysers show a zero level.

[Jester]

Ooops and not to mention the TG minimum frequency is 100kHz

[hendorog]

Your screendump doesn't make any sense. The centre frequency is supposed to be 15MHz but your marker shows 89kHz. Perhaps you should reset the spectrum analyser and test again.

It's a log frequency scale - the 'centre' value is 1.5 divisions in from the Right edge of the screen!

I missed that! But appearantly so did to OP because it would mean the left part of the screen is in the Hz range which is where (AFAIK) all spectrum analysers show a zero level.

Yep, you passed the test - you are human

It's still within spec for the SA at 90kHz at the marker (spec says 9kHz) so I'm thinking the LO feedthrough shouldn't take over at that point.

However google says that that a Siglent TG starts at 100kHz, so just out of spec for that.
https://mediacdn.eu/m/media/wysiwyg/siglent/Downloads/DataSheet/SSA3000X_DataSheet.pdf

The 'stair-step' looks a bit odd too. Like it is operating at the limit of the ADC, perhaps the atten/gain settings are not sensible for this measurement? Hit preset and try again?

[TheSteve]

I love that the listing says it is good from DC-6 GHz  - yet it blocks DC.

[Jay_Diddy_B]

Hi,

The frequency response is given by:

F = 1/2 x PI x RC

If R= 50 Ohms (Spectrum analyzer input)

And C = 100pF

The F (-3dB point) = 31.8 MHz

This is consistent with your measurements.

If you use the same DC block on a scope with a 1M Ohm input

The -3dB frequency is 1.59kHz

I assume that in this context DC-6GHz means low frequencies to 6Ghz.

But for some applications, like EMC measurements, a higher value capacitor is needed.

22nF will give you a -3dB point at 150kHz with a 50 Ohm SA.

Regards,

Jay_Diddy_B

[Jester]

Hi,

The frequency response is given by:

F = 1/2 x PI x RC

If R= 50 Ohms (Spectrum analyzer input)

And C = 100pF

The F (-3dB point) = 31.8 MHz

This is consistent with your measurements.

If you use the same DC block on a scope with a 1M Ohm input

The -3dB frequency is 1.59kHz

I assume that in this context DC-6GHz means low frequencies to 6Ghz.

But for some applications, like EMC measurements, a higher value capacitor is needed.

22nF will give you a -3dB point at 150kHz with a 50 Ohm SA.

Regards,

Jay_Diddy_B

J-D,

Your analysis is correct, looks like I need to find a DC block with a larger capacitor for conducted emissions testing.

Updated graph with 50 scope termination as well as TG/SA from 100k to 100MHz

[Performa01]

... looks like I need to find a DC block with a larger capacitor for conducted emissions testing.

I use this one from Mini Circuits and can highly recommend it:

https://www.minicircuits.com/pdfs/BLK-89+.pdf


Even though it’s rated from 100kHz to 8GHz, for most practical purposes it actually starts much earlier:



This means -0.5dB error at 15kHz.

It has SMA connectors though.
Mini Circuits also have an N-type DC-Block, but that also is 10MHz to 6GHz, hence very similar to the one you have.

[Jester]

... looks like I need to find a DC block with a larger capacitor for conducted emissions testing.

I use this one from Mini Circuits and can highly recommend it:

https://www.minicircuits.com/pdfs/BLK-89+.pdf


Even though it’s rated from 100kHz to 8GHz, for most practical purposes it actually starts much earlier:



This means -0.5dB error at 15kHz.

It has SMA connectors though.
Mini Circuits also have an N-type DC-Block, but that also is 10MHz to 6GHz, hence very similar to the one you have.

Thanks for that, I found that exact one when I searched after J-D's response.