Sir Anton . . . . .
Whut ! . . . no one's gonna' come forth, step up . . . and helps you out further, with that petite little "Jewel Box" Emerson receiver ? ?
The loop is supposed to be hidden inside of the top cover , and I enclose a pic of that frontal profile:

Now, if the original is just flat missing . . . wonder wha' hoppen' ?
I too, also thought that there was some info relevant to a loop antennas construction on the info on the home page references of this site, but did not readily run across it.
I do remember helping another person who was totally missing his loop on a Bendix table model a couple of years ago and was able to direct a procedure to him on the fabrication of one and get him in on the flight path and his having total success in completing it.
I certainly thought that I had the info still available on this sites search archives, but the site must have dropped the info.
I remember a Setchel Carlson "Frog Eyes" of mine that has the absolutely very best little "fingered" loop antenna construction that I have ever run across, so I pulled it down to take dimensions and construction technique to plot out a drawing of its physical profile.
There's assuredly nothing skimpy about its construction with a full 17 tabs, being used for intertwining the single individual windings in between.
I plotted just the dimensions on one layout and on the other, I detailed the coil winding procedure.
That unit of yours is probably being the smallest . . . . unit made, until the "peanut" and "pencil" profile . . . (even smaller tubes), came into use for making a couple of series of mini portable sets.
Now my Setch C loop turned out to measure 4.5 high by 10.5 in wide , and in referring to the provided photo for the height dimension of a loop that would fit behind that black Emerson cover and clear ~1/8---1/4 in, all around, on all the sides, I come up with a 3.5 to 10.5 RATIO of a height to width requirement for your loop to fit.
Now that 3.5 figure is a given, BUT there is a compressive parallax error in trying to compute the exact proper WIDTH figure, due to the cameras offsetting angle used in the taking of the photo.
Using my given 3.5 figure I have now skewed the dimensions of my prepared drawing to be correct for your set.
All you will have to do is print out the pattern profile initially as a full size, and then compare your height of the available space within the cabinet to the height of the print out to see mathematically what degree of size reduction in the next print size is THEN needed to make your second print out, be in agreement, size wise.
Then you cut out and trim the drawing and confirm its proper fit within the set.
I transported my hi density bit mapped pattern data into ADOBE . . . . Pee Dee Eff . . . . Formatting, so that you can be able to preeeeeeeecisely adjust reduction by 1% increments.
( That referencing URL . . . is being placed at the very end of this posting.)
HOW TO . . . DO IT TO IT:
With that sets Lilliputian size, I am expecting its RF tuning condenser section to probably be a 365 uufd unit.
The SC used #26 single cotton covered solid copper wire and 12 turns were used in the winding of the loop. Soooo 36 feet of wire would be more that adequate for my unit, so additionally use that as the worst case for the requirement of yours.
As for wire, I would be using what I have, and the only things that come to mind are a 500 ft roll of common telephone wire that I have OR now a days, myself . . or others are also apt to have some CAT-5 or its variants that would be a good source if slit open and stripped out.
Also I have some rolls of Kynar wire wrap wire, and if one has nuttin' available on hand at all, there is also Ratty Schacckamus with its small rolls of "enamel" (Formvar) magnet wire being shelf stocked items, with its #26 gauge being a quite optimal size .
You can also figure out for yourself, what materiel you want to make the form from, nothing too laterally FLIMSY however, nor too overly thick, to interfere with your being able to get the tops black cover back on.
If having to replicate my larger SC unit, I certainly would be using some leftover black Formica countertop materiel, which is about 1/16 in thick . . and very rigid..
WINDING IT:
Once you have the antenna form cut out, lets do an electrically operational analysis of the rough amount of loop inductance requirement.
Place the flat plane of the loop upon something ? that will support the loop form about 1 1/2 in above the bench.
The you get yourself an old "vanilla " file folder, or some cardboard /poster board stock of about that thickness and/or rigidity.
You should only need about a 2 1/2----3 in long strip, so cut some from the long length of the unfolded folder.
Flop the edge of the strip down on the bench and encircle the outermost periphery of the loop pattern template.
Using multiple strips being taped together is OK, as you will result in having an even more stable form.
Scotch tape the form all together and lift up , off from the pattern, to get your resulting big circle, then use tape cross pieces across the center and 1/3 in from each end to get the circular form to re-assume that same oval shaping as the templates pattern had.
Now you allow 1 foot of wire for connecting to the radio and then tape tack the wire to the loop for retention of that loose end.
Then you just rough wind on ~20 turns in the center of that new form.
Retain the new resultant wire end with tape or a big blob of modeling clay, etc.
Then you strip the wire ends and connect into the radio, get the unit working and see if you can pick up the strongest station around, which I suspect you will still be able to do with the current, probable excessive inductance.
Then you ferret out your WEAKEST receivable station and tune right onto it, and start slowly taking a turn off from the outer winding of the loop, its reception should be improving, as you take off one or a few turns.
(As you take off the wire keep the removed excess going outward as an incrementing, U shaped "hairpin" , in order to minimize its stray . . unwanted . . inductive effect.)
Now for a potential shortcut . . . I'm initially making the assumption that the trimmer capacitors on the size of the variable tuning condenser are still set at "factory adjustments".
There will be the OSC and the RF sections of the tuning condenser's trimmer capacitors and we are interested in the RF section, which should be having MORE plates on it on this vintage of a set.
Or ever in doubt on a set just tune in a station and put your fat finger so it is touching the fixed plate and the radio should shift off from the station when touching the OSC section . . . that side we are not interested in so the RF section identifies itself by default !
Next you will hopefully find a standard slotted screw use for compressing the leaf of that RF section trimmer, so place a Sharpie referencing line to the adjunct frame to one edge of the screw slot.
Now if you don't turn that screw more than 180 degrees you can find its home markings again ! or else . .you will have to COUNT your turns.
Now you will be more precisely evaluating your decreasing of loops turns effect, by swinging that RF trimmer 180 degrees to the left . . . and then back home . . . and then 180 degrees to the right.. . . and then back to its home position.
I am suspecting that the CCW action of decreasing the RF trimmer capacitance has made that weak station come in even stronger, which is indicating that the loop inductance is too high, and still needs another turn or so off, to decrease its effective inductance even more.
When you get it to the condition where you can now pick up a station in the proximity of 1400 kill-o-hurts and then rotate the trimmer and have it peak at the initial marked up position , you are right into the ball park of having the correct inductive value of your new loop.
FINALIZING IT:
You are ready now to unwind all of the wire from the test form and transfer it to the dedicated form which you have cut out.
If you consult the drawing below, you will see the use of a RED color coding of the wire at the start where it goes over and into Slot 17 and then behind Tab 17 Prime and then over the front of Tab 1 Prime and then behind Tab 2 Prime and then in front of Tab 3 prime . . .ya-da-ya-da-yah . . . in encircling around the loop profile for the making that first tight fitting turn.

Now if you have done any downhill slalom skiing . . . it's winding will be a piece of cake . . . furthermore . . . if your grandmother happened to have been a Hopi basket weaver . . . you have it aced . . . as its ingrained in your genes.
For clarity I have changed the coloring of that 1st turn, just as it replicates as being the second turn, with it being shown in GREEN for that winding portion, and then it just makes a drop out at the top, for you to go on and finish.
I have no idea as how the tab/slot ending of the final turns will be on the FUCHSIA ending turn, but it is being routed over to the terminal area . . . or to a brass/copper hollow rivet as shown.
Replicate the above inductance trimming procedure for the REAL loop antenna now.
Then ONLY you will know this, but I would expect a ribbon cable to convey the two connections from the top cover to the radio proper, via the hinge bend of the top. It might still be all intact or there might be a remnant left.
Wire would not be desirable at all, so if you have to make that up completely, get some thin brass shim stock from a FULL SERVICE HARDWARE store.
Its offered in rolls about 3 in wide and they cut off the inches / foot that you need.
Then you cut two flexible brass strips about 1/16 in wide and 1/8 an inch apart with clear plastic sealing tape above and below the strips and then trimmed down, that will make you a custom, flexible ribbon cable of your desired / required length.
Now as for the template of the loop form . . . . with the bad news being that neither Image Shack nor Boto Phucket seem to now be accepting my PDF format of an image ?
Some time back, I successfully prepared up a blow up and Pee Dee Eff'ed up Sir Marv's Rat Shack VOM meter and it was fully accepted by Image Shack hosting back then.
But just maybe, that filing of PDF formatting is a pweemium pwiced option now ?
Any whooo ? . . . here it is . . . being presented as a fine detail PNG and if you will then open it ups with Winders " Pitcher and Facts Viewer" it should be possible to scale it down there, with its capabilities along with the options on a printers settings.
(Do make a liberal utilization of "Print Preview")
Thassit . . . . . unless you have further info queries . . . . . .

URL access to the above drawing template is:
http://img690.imageshack.us/img690/6273/looptemplate.png>
( For any interested viewers up in the peanut gallery, run off your template copy now, for any potential future use.)
73's de Edd
