SBL1 Balanced Mixer Ideas ?

[pwdrive]

Has any one got any thoughts of what would be the best way (if it's possible) to match the three ports of the SBL1 Balanced Mixer (or similar device) to valve circuitry, the application would be for a Direct Conversion Receiver, Ideally with an RF amp preselection stage.

Lawrence.

EDIT:
Forgot to add: I think this device is 50 ohms on all ports....

[Skywave]

Lawrence - can you post a link to where the spec. of the SBL-1 device is - or any others that you are considering?
Thanks - Al.

[pwdrive]

Hello Al, thanks for the reply, this is the only data sheet I can find:

http://www.minicircuits.com/pdfs/SBL-1-1+.pdf

I remember experimenting with one of these years ago in a homebrew direct conversion receiver but that receiver was a solid state job, it was a long time ago and to be honest I can't remember what was what only to say that it worked and worked quite well.
I am wondering if the SBL1 (or similar type) can be grafted into a valve direct conversion receiver. I could not find the port impedances on the data sheet but I am pretty sure that they would be designed to match a nominal 50 ohms. I am thinking that the A/F output impedance from the device could be stepped up to feed a valve A/F stage by what ever means and that a VFO feed could be arranged via a suitable winding or lower end tap on the VFO coil.
So far as the R/F in is concerned then an L/C matching network between the aerial and input port would probably do but I would like to build a valve R/F preselection stage if it is possible, I am guessing that the anode circuit of that would have to be broadly tuned if I was to step down the anode impedance down to 50 ohm for the input port of the SBL1, the grid circuit would be standard parallel tuned with an aerial coupling winding to give the old time 400 ohm (approx) aerial input impedance. I realize that there might be a few small mismatches along the way but this homebrew job would be for Rx only not Rx/Tx (as in QRP) Any thoughts would be appreciated bearing in mind I'm not an R/F designer.
Thanks.

Lawrence.

[pwdrive]

Hello again Al, I thought I would post this up:

http://www.epanorama.net/sff/Radio/Commu...rt%201.pdf

Down at the bottom of that PDF page is a simple direct conversion receiver using a single balanced mixer with a balance control, it was this article which started off my interest in that type of receiver back in the 70' I built that one but used vlaves instead, I seem to recollect that I used an EF183 or EF184 for an added RF stage, an EB91 for the two mixer diodes and an ECL82 for the A/F stage, the coil was triflar wound (mine was on a bog roll) I did not build a VFO, instead the injection input was a straight feed from an old Marconi TF144 standard signal generator I had at the time. I seem to recollect that this circuit would null out any strong out of band AM breakthrough when resolving CW/SSB, it too seemed to perform quite well considering it's relatively simple design.
Part two of that article is also available on the web, just Google.... direct conversion pat hawker.....

Lawrence.

[Skywave]

Hi Lawrence,
I haven't forgotten or ignored all that. There's a lot going on in my life right now, and there's a lot of stuff in those references to digest. When I do come back, I want to be sure that anything I do say is helpful / not rubbish / constructive: this thread is in my 'In Tray'. All I will say for now is that I'm impressed by the price.

Al. / March 3, 2012.

[pwdrive]

Thanks Al, no worries, I don't think the Mini-Circuits SBL1 is in production anymore but some NOS are still available, the nearest Mini-Circuits double balanced mixer that is still in production that I found is their ADE 1L, these are not expensive and are more or less the same spec as the SBL1 but with slightly less insertion loss. As regards matching the 50 ohm ports to valves I was thinking maybe for the A/F out port a simple A/F step up transformer to the cathode input of a simple grounded grid cascade twin triode, this would make the turns ratio on a homewound matching transformer a lot lower, I think maybe matching the 50 ohm R/F in port and Osc in port to the R/F amp and Osc might be reasonably straight forward, the idea of an R/F amp is to overcome to a certain extent the insertion loss of the passive mixer, Mini-Circuits do active versions of their double balanced mixers ie: buffered, but I am trying to keep whatever I eventually build to valves and diodes only. From what I can gather the impedance matching of the output port is the most criticall in preventing the mixer being out of balance ie: R/F and Osc breakthrough.

Lawrence.

[Skywave]

I see that the RF and LO inputs are not balanced to ground whereas the IF output is. And I would say that you are most likely correct in assuming that the source and load impedances are 50 ohms (This, of course can be determined by measurement).
What immediately springs to mind is a 1:1 toroid on the LO and RF inputs with each of the primaries being driven from a cathode follower (Zout = 50 ohms) - or a higher figure and change the turns ratio of the toroid accordingly - but I wouldn't exceed an impedance transformation of about 3:1. As for the IF output, a roofing filter might be advisable or perhaps straight into the primary of the 1st. IF transformer. Obviously, a 50 ohm match will be required here and the drive level of the LO is important - but you are doubtless aware of that.

Aside: In the past, when I've worked / experimented with DBMs, I've always found a spectrum analyzer to be an essential tool.

When I find the time to think about this a bit more, I'll come back.

Edit: this pdf from Mini-Circuits: [attachment=4204].
Look at the last entry: it makes it clear that Mini-Circuits DBMs are designed for 50-ohm systems.

Al. / March 3, 2012 //

[pwdrive]

Hello Al, thanks for that, just to mention that the IF out from the mixer will actually be base band audio ie: Fo (Osc) =
Fs (Signal) resultant = audio. Not knowing to much about these things can I ask why you suggest 3 to 1 max. ratio on any matching transformer for the Osc and R/F ports.

Lawrence.

[Skywave]

A.F. as output: yes, O.K., fair enough! I always think of DBMs as RF in and RF out devices: I hadn't considered mixing to baseband; can't see why that should be a problem though.
As for the 3:1, I just have this notion that a ratio greater than that may introduce problems of bandwidth limitation, on account of excess distributed capacitance in the winding with the greater number of turns & capacitive coupling, pri. to sec. But then again, I may be talking through my hat! (not unusual ) So a bit of development work may be required (which as I implied earlier, is a lot easier with DBMs if you have a spectrum analyser - and a couple of good sig. gens.)

Al.

[pwdrive]

Thanks again Al, yes the mixer will be the basis for a direct conversion receiver, according to the specs the output from the mixer is good from DC up to around 400Mhz if all properly terminated so audio won't be a problem, as you say a spectrum analyzer would be a help but unfortunately I am not in posession of such an instrument, all I have is a cheapo DMM, an old Avo Mk7, an Avo Mk1 sig. gen and an Avo CT 378A sig. gen. plus my ears......A bit of experimentation I think, it probably won't be 'till next winter because I have a lot of house DIY to see to, next project is to empty and desludge the septic tank and convert it to a full two stage job with a seperate sump/pump chamber to pump the semi treated effluent up to my new land drains (land drains higher than septic tank) then its converting house roof from corrugated sheet to tiles so I will be busy....Having said that I intend gathering all the bits I need for the direct conversion receiver before next winter so that I will have all the bits ready to start the project. I understand what you say about a large transformer ratio on the Osc and RF ports and will take that into consideration.

Lawrence.