EMRFD Message Archive 9959

Message Date From Subject
9959 2014-04-27 10:31:55 David C. Partridg... Question about RF amp ICs and MMICs
Hi folks,

My name is Dave Partridge.

This is my first post here, and I've almost not RF experience apart from reverse engineering the levelling head for the Tektronix SG504 oscilloscope calibration generator. See:

I'm mulling over the possibility of replicating the entire function of the Tektronix SG5030 and/or SG5050 levelled sine wave generators (not just the levelling head) using modern technology (tho' it must be said that these aren't that old tech). The original SG5030 produces outputs from 0.1Hz to 550MHz, and it's bigger brother (the SG5050) goes up to 2.5GHz. Output levels from 4.50mVpp to 5.50Vpp into 50R (-42.95dBm to +18.75dBm).

Leaving aside the issue of the DDS and PLLs for now, I'm looking at the output amp and wondering if its function could be replicated with a MMIC (or similar). My problem is that while most MMICs claim DC-nGHz, in practice they can't operate at DC as they seem to require an input coupling capacitor and get power using what's effectively a bias-T at the output terminal so they are AC coupled.

My question is do any of the RF gurus here present know of any MMICs (or other RF ICs) that actually can be operated all the way from DC to daylight (whatever daylight is for that part) and can happily produce about +25dBm or so before compression sets in. DC-3GHz (or maybe a bit more) will do fine.

I realise I *could* dual path this but that adds to the cost.

Many thanks
9966 2014-04-28 06:12:53 biastee Re: Question about RF amp ICs and MMICs
Hi Dave,
The bias-tee is optional. You can replace the choke with a resistor and a higher supply voltage. The input cap is a little more difficult to eliminate because it requires the driving signal to have no DC offset, or else saturation will ensue.

The MSA-0520 (the BeO packaged version of the MAR-5) has the following typical specs: P1dB = 25 dBm. BW = 2.8 GHz.
73, Chin-Leong Lim, 9W2LC.
9970 2014-04-28 09:30:18 David C. Partridg... Re: Question about RF amp ICs and MMICs
There's nothing in the MSA-0520 datasheet to suggest that either of the input or output blocking capacitors are other than necessary.

I'm pretty sure that if you check the DC level at the input in a typical circuit it will *not* be at zero volts.

David Partridge


9971 2014-04-28 10:04:54 David Re: Question about RF amp ICs and MMICs
The input and output DC levels are *not* at zero volts which is easy enough to
see from the schematic of the typical Darlington configured shunt feedback MMIC
amplifier but that does not preclude using them without DC blocking capacitors.

The solution as far as the output is concerned is to offset the MMIC amplifier
ground in the negative direction to DC balance the output. The input is then
either driven with an already DC balanced signal or more likely a current
through a shunt resistance to the new offset ground.

The SG5030 hybrid almost certainly does something like this with a
simplification; it has separate high and low frequency inputs with a DC blocking
capacitor on the high frequency input. The output DC balance is controlled by
the low frequency DC coupled input. For low frequency operation down to the 0.1
Hz specification, only the DC coupled low frequency input is used. It is worth
noting that the SG5030 output *is* AC coupled by the output low pass distortion
filters in all but low frequency mode.

That should be easy to duplicate with standard MMIC amplifiers. The high
frequency input would be AC coupled like with the SG5030 hybrid and the low
frequency input would drive the MMIC ground to both DC balance the output and
provide low frequency operation. A shunt or series regulator referenced to the
offset MMIC ground would provide output biasing. Since the MIMIC is not passing
DC that could be via an inductor but a resistor would be better for wide
operating frequencies even though it sacrifices performance.

Ultimately that is just another parallel path wideband amplifier but running DC
through the MMIC RF input is probably more difficult to do without oscillation.

On Mon, 28 Apr 2014 15:54:56 +0100, you wrote:

>There's nothing in the MSA-0520 datasheet to suggest that either of the input or output blocking capacitors are other than necessary.
>I'm pretty sure that if you check the DC level at the input in a typical circuit it will *not* be at zero volts.
>David Partridge
9972 2014-04-28 10:18:32 Dana Myers Re: Question about RF amp ICs and MMICs
9973 2014-04-28 10:38:33 David Re: Question about RF amp ICs and MMICs
On Mon, 28 Apr 2014 10:18:26 -0700, you wrote:

9974 2014-04-28 13:05:40 David C. Partridg... Re: Question about RF amp ICs and MMICs
My target is 0.1Hz



9975 2014-04-28 13:35:14 n2msqrp Re: Question about RF amp ICs and MMICs
Have you looked at using a DC coupled Video Op-Amp with a plus and minus split power supply?

Mike N2MS
----- Original Message -----
9976 2014-04-28 14:05:17 Ephemeral Re: Question about RF amp ICs and MMICs
This sounds like an application for a DC offset nulling feedback loop. You will need to roll off
the feedback at 0.1 Hz.

CTRL + Click to follow link" target="_blank" href="mailto:emrfd@yahoogroups.com">emrfd@yahoogroups.com
Subject: RE: [emrfd] Re: Question about RF amp ICs and MMICs


My target is 0.1Hz



From: emrfd@yahoogroups.com [mailto:emrfd@yahoogroups.com] On Behalf Of Dana
Sent: 28 April 2014 18:18
To: emrfd@yahoogroups.com
Subject: Re: [emrfd] Re: Question about RF amp ICs and MMICs

How close to DC must the response be? Perhaps a larger DC block capacitor is
adequate. Attempting to produce a temperature stable DC offset like this
like it might prove more challenging in practice than theory.

Dana K6JQ

9977 2014-04-28 14:19:56 Jochens Feldhaar Re: Question about RF amp ICs and MMICs

I think that the solution to this problem is the combination of a RF
amplifier and a low frequency DC coupled amplifier controlled by an
OP-AMP. The trick is in combining the two components to form a clean
crossover between the two amplifiers. A lot of oscilloscope
preamplifiers uses the principle, just go and see what of these
schematics is usable for the project.

Regards, Jochen DH6FAZ

Am 28.04.2014 21:55, schrieb David C. Partridge:
> My target is 0.1Hz
> Dave
> ________________________________
9978 2014-04-29 03:17:14 David C. Partridg... Re: Question about RF amp ICs and MMICs
I was so hoping to avoid that as it adds to the cost quite a bit. Ho-hum

-----Original Message-----
9994 2014-04-30 19:43:59 David Nushardt Re: Question about RF amp ICs and MMICs
Hi Dave:Welcome to EMRFD, Small world! Nice to see you post here, Don't know if you have a copy of EMRFD but this group is based on that book , which is about experimenting with QRP circuits, IF your not a Amateur Radio Op, QRP means low power, generally under 5 watts but could be anything below about 80 watts, it also stresses high preformance receiver design.

When I was recovering from my first stroke , rehabing in the nursing home a Navy friend lent me his handitalkie so I could keep in touch with some of my ham friends, at first I was afraid, loosing the use of my left arm and hand would end my life long electronics hobby, I happen to buy a copy of EMRFD to give me something other than my condition to think about, EMRFD introduced me to the ugly construction method, that gave me new hope, and something to look forward to. I been doing exceptionally well ever since, even taking the leukemia diagnosis in stride.

By the way one of the authors of EMRFD W7ZOI is a former Tektronix engineer, you'll notice much of the circuits use the same simple transistors and common parts,  like tek gear. small world!

IF you don't have EMRFD you can get it for around 50 USD through ARRL american radio relay league.

Funny your thinking about using modern parts to duplicate the SG Generators, I'm not privy to the manuals, I too recently discovered and am building a DDS signal Gen and have been thinking if it would be possible to build something that would approch the preformance of the Hp 8640B using modern

For the present I working on building a swept DDS based signal gen with built in didital attenuator and AD based power meter, all be it very slowly! check out SV1AFN's web site for ideas!

For me this stuff has become more about the challenge of building than the engineering, to appreaciate what i mean is, try winding a Torroid core with one hand, you'll have to find novel ways to do it first!( that is the hardest for me) it's more of a exercise in patience than anything!

I have yet to finish your leveling head, I have my own Aoyue 968A and a cheap microscope camera from Amazon now, i keep getting side tracked! I have everything needed to finish including a spare normalizer for the housing , except the dang Lemo connector, can you refresh me on the correct part numbers for the Lemo? I need to get back to it soon.

By the way I have 2-SG504's one has a trashed front and sub panel, there both good I check them with a DC508A and they both read out nicely with the 508A, had a thought of using the tuned cavity as the basis for a signal gen, but I'm pushing 60 and it's time to pass my tek gear down to someone who will use it, while I'm still able.

Time to move on to other things, my wife is giving me more and more static lately just having a hard time to find the energy to pack it up, so if you need anything or know someone who is , i will be selling off my collection of Tm500 gear.

Best of luck on your SG project, please keep me posted on your progress!

Best Regards