Low level generators?

[PartialDischarge]

I'm searching for a generator (sinewave is enough) in the 0Hz to at least 1 or 2MHz, that starts at low levels, like -70 -80dbm.

Noticed the HP 3336C, any others I should be aware of?

Thanks

[forrestc]

I'm searching for a generator (sinewave is enough) in the 0Hz to at least 1 or 2MHz, that starts at low levels, like -70 -80dbm.

Noticed the HP 3336C, any others I should be aware of?

What type of application are you using this for?   I'm assuming since you're specifying the output in dBm you're looking for something specifically for a RF application.   Mainly I'm wondering about signal purity, etc?

I have a 33120A Arb function generator which does up to 15Mhz, and goes down to -22dBm.   Don't discount the use of rf attenuators to knock down the signal from even 0dBm to -80 or so.   If you're willing to put a decent amount of attenuation in the path, my experience is that many signal generators will work for a RF source.

[PartialDischarge]

What type of application are you using this for?   

Test of low noise, high gain amplifiers with inputs in the 50-100uV range at low freqs. Not really RF but what I've seen out there (HP synths) are speced in dbm.

And I have a nice SDG2042 whose output I could attenuate if it is clean enough...maybe should look into this

[PartialDischarge]

Don't discount the use of rf attenuators to knock down the signal from even 0dBm to -80 or so. 

So maybe I should get an HP 355D attenuator and connect it to the SDG2042..

[tautech]

What type of application are you using this for?   

Test of low noise, high gain amplifiers with inputs in the 50-100uV range at low freqs. Not really RF but what I've seen out there (HP synths) are speced in dbm.

And I have a nice SDG2042 whose output I could attenuate if it is clean enough...maybe should look into this

-50 dBm is the minimum setting so -70 should be achievable with additional external attenuation but be aware the output need be set to expect a 50 ohm load to have dBm available as a unit of amplitude adjustment.

[forrestc]

Don't discount the use of rf attenuators to knock down the signal from even 0dBm to -80 or so. 

So maybe I should get an HP 355D attenuator and connect it to the SDG2042..

Looks like the 355D is good up to 1Ghz, so yeah, that might work.  I didn't even know this existed, so I might have to pick one of these up.   

Currently, I have a set of inline attenuators which are good to up to 10Ghz I think.   I think the biggest is 40dB, and I think I have like 3-4 of those, and then I have various smaller steps (20dB, 10dB, 5dB, 2dB, 1dB), so I can stack a reasonable number of them to get the exact attenuation I want.   I've been known to use them to directly connect two 2.4 or 5.7Ghz radios together and simulate free space path loss on the bench.   

[DaJMasta]

Inline attenuators are the easy way to do it, for such low frequencies you could probably build your own and coarsely verify them with a scope FFT or something.  You could also buy a used mechanical attenuator switch - many are pretty simple to drive, just some basic logic level binary to drive the stuff built in that drives the solenoids.  That way you could get a little simpler variability and a little more repeatability - though that is only likely to be noticeable at higher frequencies.

[nctnico]

Or get this relatively cheap (US $80) manually operated step attenuator:
https://www.ebay.com/itm/RF-Coaxial-Adjustable-Step-1db-Attenuator-SMA-connector-DC-2-5GHz-5-Watts-0-90dB/360887349827?hash=item54068ff243:g:XYUAAOSwq1JZGRu3



When I need low levels I use this or fixed attenuators. The limitation will be common mode noise from the generator.

[PartialDischarge]

Thanks to all. I'll try with these I just bought from epay, $42 both+shipping, they surely need a bit of soap and a toothbrush

[MadTux]

HP 3325A,3326A, 3335A and 3336 (don't know that one personally). HP 3335A and especially HP3326A are built beautifully.
Maybe Tek SG505+attenuator, if you don't mind the crazy price for a small TM500 plugin 
Tracking output from HP3585A is also nice for sweeping, but requires external programmable attenuator box.

[Fungus]

Use a potentiometer to attenuate the output of an existing generator...?

[PartialDischarge]

Use a potentiometer to attenuate the output of an existing generator...?

Well, I’m not that cheap and prefer to have something with a defined frequency response, att ratio, shielding... ie for $42 I can’t ask for more with those hp 355

[tggzzz]

Thanks to all. I'll try with these I just bought from epay, $42 both+shipping, they surely need a bit of soap and a toothbrush

Those HP355x will be just fine - presuming they work to spec, of course!

Don't forget the indicated attenuation presumes a 50ohm load.

[CopperCone]

Most function generators assume a 50ohm load so its not really changing much.

[tggzzz]

Most function generators assume a 50ohm load so its not really changing much.

The generator isn't the issue! If you omit the 50ohm load on the output of the attenuators then the voltage will be 3dB / twice higher.

[PartialDischarge]

Well, I received the HP 355D and 355C. Of course I cleaned them, before



After cleaning,




The inside seems clean, no burned resistors or bodged repairs,




Notice the crimp between the resistors and the body metal,



The center pin of the BNC connectors doesn't seem abused,




Haven't tested them yet, but I expect no problems, so for $42 both I'd say it was a good buy