Fillet Question

[rwanttaja]

I'm designing a small enclosure for a headset matching adaptor. It's a small bracket that holds the 1/4" jack and the matching transformer, and an exterior cover slides over it.

The bracket features a shelf where the transformer attaches (flanges on the end of the transformer go into the little holes in the shelf and are bent over to hold it), and a hole in the end for a 1/4" jack.

http://www.wanttaja.com/fillet.jpg

When 3-D printing this, I've encountered a structural issue at the point where the shelf meets the main structure. I'd like to add a fillet to the area pointed out by the yellow arrow. Yet I can't get SW to understand what I want to do.

Any advice? I could make the shell go all the way the end, but that would complicate installation of the electronics.

Ron Wanttaja

[cwilliamrose]

I wish I had seen this while I was still at the shop. But just looking at it I don't see why it's not a straight forward fillet. You should be able to select the intersecting line and create the fillet. Sometimes SWx gets fussy about the fillet radius but this doesn't look to be that type of a problem. In any case once the intersecting line is selected you can play with changing the radius (start small). Also, try just one fillet at a time until you get something that is acceptable, then try adding other intersecting lines to the selection set. Sometimes SWx forces you to use multiple fillet features to get all of them you need into the model.

Edit: This is a single body part, right?

[rwanttaja]

I wish I had seen this while I was still at the shop. But just looking at it I don't see why it's not a straight forward fillet. You should be able to select the intersecting line and create the fillet. Sometimes SWx gets fussy about the fillet radius but this doesn't look to be that type of a problem. In any case once the intersecting line is selected you can play with changing the radius (start small). Also, try just one fillet at a time until you get something that is acceptable, then try adding other intersecting lines to the selection set. Sometimes SWx forces you to use multiple fillet features to get all of them you need into the model.
Sigh (and slap my head).

I'd been selecting the aft face of the oval area, AND the side of the table area, or the vertical line AND the aft face of the oval area. Never considered just clicking the vertical line only.

And yes, that did the trick. Thanks!

Just got done printing the fillet-less version on the 3D printer. If it breaks coming off AGAIN, I'll do the filleted version.

Ron Wanttaja

[gmatejcek]

Or, one COULD do like the old kits, and use a plastic 35mm film canister... um... nevermind!

[rwanttaja]

Or, one COULD do like the old kits, and use a plastic 35mm film canister... um... nevermind!

Yeah, they're getting kind of thin on the ground. The very first headset adaptor I made used a 35mm film canister.

I'm tickled with idea that I can make custom enclosures, though, using SW and my 3D printer. Not only can size them right, but I can build in compartments to hold components like the transformer. This unit worked (tested it in the airplane this afternoon) but it's still bigger and clunkier than I want. Working on the follow-on.

Ron Wanttaja

[cwilliamrose]

Did you need those fillets Ron?

[rwanttaja]

Did you need those fillets Ron?

I was able to extract the piece from the 3D printer bed without them, but obviously, I'll want to include fillets on any future tries. Thanks for kicking me in the head, on this one.

The next iteration will be a fairly different design. I've drawn both types of transformers in SW, and am working in the assembly mode to get used to putting things together. Takes a bit of practice....

Ron Wanttaja

[cwilliamrose]

Yeah, mates can be confounding sometimes. I use a 3D mouse so it's easy to move a part into a nearly correct location before applying the mates.

[rwanttaja]

And...as can be seen, I learned my lesson...both literally and liberally:
http://www.wanttaja.com/adapt.jpg
This is a different design from the earlier posting.

The blue item is a low-detail 1/4" headset plug, red thing is the matching transformer, and the gold bit is the 1/4" headset jack. The transformer is shown "tabs up" because that's the end of the transformer where the wires come out. I'm intending to pot the thing is place with epoxy. A tubular outer shell will be slid into place, and when the unit is done, it'll match the impedance for my ANL aircraft headset to my 8-ohm radio.

The plug is a standard unit; I unscrewed the black plastic covering over the connections, measured the threads, and reproduced them on the unit using the SW hole function. Screws right in. Probably will epoxy it to be on the safe side.

As expected, my 3D printer had problems with this....it doesn't like to print unsupported shelves like the transformer support and the end piece with the headset jack. I've generated a new version where the end piece is separate, and will just epoxy it in place.

The ironic thing here is that I laid out the components on my desk and measured what the total length would have to be: 2 1/4 inch. But when I put it into SW and built the assembly, 2 1/4" was too short... stuff didn't fit. Didn't believe it...but did a 2 1/2" version anyway. Everything (barely) fits, just like Solidworks shows.

I know there are lots of guys doing things far more complex, but this is turning out to be a good learning experience. Now to see if I can renew my SW license....

Ron Wanttaja

[cwilliamrose]

Looks good Ron. One thing I see is the tabs on the transformer being red on the outside face. You could make those separate bodies in the model so they could be silver/gray on all sides. Or you could take a .0005" cut on the outside faces so there's a separate face to match colors with. This is only important if the details of the artwork need to be accurate.

Almost looks like the headset jack couldn't be assembled if the housing was all one part......

[rwanttaja]

Almost looks like the headset jack couldn't be assembled if the housing was all one part......
Was preying on my mind, too. :-)

Rotating the jack to put the contact stanchions on top seems to leave clearance for the jack to lever into place. It can then be turned to the desired angle. The printer doesn't generate the transformer-bay bulkheads very well (they're more crescent-shape) and that leaves more room.

I'm not that fired about this design because it cantilevers off the instrument panel; don't like the way loads are going to go. I'm working on a more-squat design.

Ron Wanttaja

[flyboy2160]

...I'd been selecting the aft face of the oval area, AND the side of the table area, or the vertical line AND the aft face of the oval area. Never considered just clicking the vertical line only...

Sometimes using the standard 'edge fillet' method of picking the edge in the crotch won't work. You can then try a 'face' fillet, which is closer in practice to your original pick scheme. You pick the faces between which you want the fillet.

[rwanttaja]

Assembled...but not wired...unit, both open, and with the outer shell in place.
http://www.wanttaja.com/adapt2.jpg
I'm thinking about making one 1/8"=1/4" longer, to make wiring a bit easier.

For those wondering, when you are using an aircraft headset on a handheld radio, this adaptor increases the headphone sound level by 6 dB (doubling the volume).

Ron Wanttaja

[Kyle Boatright]

I'm not that fired about this design because it cantilevers off the instrument panel; don't like the way loads are going to go. I'm working on a more-squat design.

Ron Wanttaja

Can you mount it behind the panel, with only the jack protruding?

[rwanttaja]

Can you mount it behind the panel, with only the jack protruding?

I've considered that... in fact, just mounting a second jack in the panel with the matching transformer. Then I could plug in to the "aircraft headset" or "ear bud" jack as appropriate. A bit more work than I want to do, unless I have to peel open the behind-panel area for another reason.

Ron Wanttaja

[flyrgreen]

[QUOTE=rwanttaja;79194Ron Wanttaja[/QUOTE]I'm not that fired about this design because it cantilevers off the instrument panel; don't like the way loads are going to go. I'm working on a more-squat design


I have worked with all things audio for many years and saw this problem right away. You are correct to be concerned about the moment-arm weight from all this hanging on the 1/4" jack. By the time the actual headset cable and plug are connected there is considerable force. I don't have a guess about this force making for actual breakage, but it will make for internal connection problems. You likely know that there are factory-made transformer adapters just like this, for sale at any audio outlet (including Radio Shack if you can find one). But they are made of metal and much heavier than your SW design. I'm guessing that's why you are bothering to do this work. But the cantilever and significant moment arm seems dicey to me. I would try to design something that is a 90 degree plug.

[rwanttaja]

I'm not that fired about this design because it cantilevers off the instrument panel; don't like the way loads are going to go. I'm working on a more-squat designI have worked with all things audio for many years and saw this problem right away. You are correct to be concerned about the moment-arm weight from all this hanging on the 1/4" jack. By the time the actual headset cable and plug are connected there is considerable force. I don't have a guess about this force making for actual breakage, but it will make for internal connection problems. You likely know that there are factory-made transformer adapters just like this, for sale at any audio outlet (including Radio Shack if you can find one). But they are made of metal and much heavier than your SW design. I'm guessing that's why you are bothering to do this work. But the cantilever and significant moment arm seems dicey to me. I would try to design something that is a 90 degree plug.

Yes, and the problem here is actually worse than it might first appear. I'm building this adaptor to get good volume from a Flightworks Denali ANL headset; as an ANL headset, it has a battery box on the cord about ~6 inches from the plugs. So the cantilevered design, if I just plug in the headset, is even worse with the weight of the battery box added.

As you can see from the aircraft installation, I do have options. There's a small shelf below the headset jacks; it's where I used to clamp the battery box from the Denalis before I decided I needed more volume.
http://www.bowersflybaby.com/stories/fun_meter.jpg
I might actually build a little clip to attach the headset's battery box to the GPS mount (the disk-shaped thing just above the ferrite cores on the headset cable). That would take most of the load off a plug-type adaptor.

I actually made a "long" version of an adaptor cable; ~3' of cable terminating in a box with the jacks. This works, but there's too much cable. Might do a short version that attaches to the Denali battery box.

This is a nice casual project. In the summer, I use 8-ohm ear buds, and in the winter, I wear a Rugged Radios headset where I've swapped out the 300-ohm speakers for 8-ohm units. No rush, no impact if it takes me a while to solve.

Closest Radio Shack to Seattle that I know of is in Skagway, Alaska...that's a fur piece to drive. I know they had some transformer adaptors before, but never saw a 300-ohm to 8-ohm version. I used a lot of Rat Shack audio transformers in my own adaptors, though.

Ron Wanttaja

[rwanttaja]

And...test fit went pretty well. Was able to fit the ANL battery box across the strap below the radio, and the adaptor actually rests on it a bit.
http://www.bowersflybaby.com/conv.jpg
I'll probably remove the ferrite cores; believe my feedback problem has another source.

One of these days, hopefully, it'll stop raining long enough to take it out for a test....

Ron Wanttaja

[flyrgreen]

Yep, that looks like good support. I had a 'feedback' sounding radio problem when I bought my (used) plane. After much poking and cursing, turns out it was a weak ground on the BNC antennae connector on the tray. I see you are handheld so no tray, but mentioning this to say that a working but very weak ground on the antennae side will choke the signal and make it cram back into the radio's output. Imagine if that were you--- you'd be be squealing too!

[rwanttaja]

Yep, that looks like good support. I had a 'feedback' sounding radio problem when I bought my (used) plane. After much poking and cursing, turns out it was a weak ground on the BNC antennae connector on the tray. I see you are handheld so no tray, but mentioning this to say that a working but very weak ground on the antennae side will choke the signal and make it cram back into the radio's output. Imagine if that were you--- you'd be be squealing too!

Yeah, I had a squealing (audio feedback) problem that was driving me nuts. No problem plugging in the headsets directly, no problem when using 8-ohm headsets. Problem ONLY occurred when an impedance matching transformer was used. Used another handheld at home for testing, but could never reproduce the issue. Thought it might have been some leakage on the antenna coupled with the windings of the transformer, but even with the adaptors wrapped in copper tape, still had the feedback.

Finally figured it out: Sidetone volume level was too high. "Sidetone" is hearing your voice in the headphones when you transmit. Adding the adaptor increased all audio in the headset by 6 db...which is *four times* the audio power. I cranked down the sidetone volume (keypad adjustment on the Icom) and the squeal went away. And I did still have sidetone at a decent volume level.

Ron Wanttaja

[flyrgreen]

So was ACTUAL acoustic feedback! Don't know if I would have thought of it. But makes sense since ANL headsets have microphones pointed outward, toward the pilot's mic.

Meant to compliment you on your FUN gauge--- I see it's permanently pegged to redline, just the way it should be.

[rwanttaja]

So was ACTUAL acoustic feedback! Don't know if I would have thought of it. But makes sense since ANL headsets have microphones pointed outward, toward the pilot's mic.
Actually, the situation was way weird. I have an Icom at home (ICA23) that I used as a test unit with the standard Icom headset adaptor and the rubber duckie antenna, and an ICA5 in the airplane with a homemade adaptor (leading to the jacks in the picture) and a coax going to the antenna.

I wear eight-ohm ear buds and homemade mike holder in warm weather, and a Rugged Radios headset that converted to 8 ohm in cooler weather. Neither has shown any feedback.

I have a Denali ANL and a cheap non-ANL headset. No squeal for either when plugged into either the airplane-mounted handheld or the separate one on the bench.

Yet when I did an adaptor, I'd get squeal with either headset in the airplane, but never on the bench unit.

My working theory was that I was getting some sort of coupling back into the transformer itself. So I did several adaptor units, wrapping them in copper tape, or using long cables so the transformer was nowhere near the radio or the antenna. None of them got rid of the feedback.

What was most frustrating is that the feedback NEVER occurred with the bench radio. Never. If it did, I would have been able to test various combinations to see that it might be....but it never showed the problem on the bench.

Until I got desperate.

I wondered about the sidetone level, so I cranked up my benchtop radio and gradually turned up the sidetone. It did, eventually, start squealing.

So I went back to the airport and repeated the process, turning DOWN the sidetone level. Sure enough, the noise finally went away, and I could still hear my voice when I transmitted.

My current thinking is that this was not air-coupled feedback, but the vibration of the headset structure itself being picked up by the microphone. Both my ANL and the cheapie headset use solid mike booms.

Adding the adaptor increases the headset volume by 6 dB (actual measurement), which is FOUR TIMES the sound energy. That might explain why it was coupling back.

But...why did the feedback NOT happen with my 8-ohm headsets? The sound level was just as high.

The Rugged Radios unit has a wire-cage mike holder, vs. the solid, flexible mike holder of the two 300-ohm units. Wonder if that attenuates the vibration better. With the ear-bud headset, there is no physical connection from the speakers to the mike...the ear buds go into my ears, and the microphone is held in place by an entirely different structure.

In any case, it was causing me lots of grief. http://www.bowersflybaby.com/pilot_nuke.gif


Meant to compliment you on your FUN gauge--- I see it's permanently pegged to redline, just the way it should be.

The picture isn't sharp enough to tell, but the needle is ACTUALLY just at the top of the yellow arc. It's a high level of fun, but not necessarily dangerous, yet.
http://www.bowersflybaby.com/stories/meter_finished.jpg

Construction instructions at:
http://www.bowersflybaby.com/stories/fun_meter.html

Ron Wanttaja

[flyrgreen]

The solid boom arms holding the mic will definitely transmit much more sound than the wire cage affair.

[rwanttaja]

Well for those who love saying "I told you so..." :-)
http://www.bowersflybaby.com/tech/broken_adaptor.jpg
But this didn't break from normal loads. I was getting out to put gas in the airplane and left my headset plugged in. Whacked the adaptor with my knee.

The external shell is broken as shown, and the glued-on end piece of the internal structure broke free. It can just be glued back into place. I could superglue the external shell, too, but I'll just go for a new one.

This is the beautiful thing about using Solidworks. If I'd built that from scratch, it'd take me a bit to duplicate, and I'd have to fiddle with the new piece so it'd match the old one. Instead, I just dig up the file and send it to the 3D printer.

Contemplated adding some thickness or some stiffeners, but I think I'll be satisfied with NOT leaving it plugged in while I climb in or out. If it breaks again, I'll look at some upgrades.

Ron Wanttaja

[flyrgreen]

An ejection seat mishap would have made a much more interesting story, so points to you for keeping it real.