Aux. air regulator woes., Engine unit dead, replacement too! Options

[redshift]

one word: remote control servo drive auto-manual HAR

(human air regulator)


M

[markb]

PM Cap'n Krusty. He has 2 working AAR's on the shelf at the shop. Dunno what he wants for 'em.

[jsteele22]

Hey Air Cooled Nut,


Your pics inspired me to rip my AAR open. Mine had been working okay, but the wire was so old & cracked the insulation didn't do its job. Like a goober, I yanked on it, thinking that that *must* be a connector in there. No, it is not. So anyway, my task was just to solder in a new wire.

I never looked at my AAR too closely. From your pics, though, I realized that the unit is sealed by a "squish" seal (I'm sure the Germans have an impressive word for it...) The bottom is a can with an embossed lip, and the top is a flat flange. When the flange is in place, a machine of some sort squishes the lip tightly against the flange.

To open, I put AAR gently in a vise, and started hitting the seam with a chisel and hammer. The chisel was digging under the lip, pointing radially outwards. Problem w/ this method is that the tubes of the AAR are in the way. But once the lip is curled up in one spot, I found that the metal is soft enough to pry up w/ a flathead screwdriver.

For my job, I drilled out the contact where the wire used to be, using a very small bit. The shavings that came up were copper. Bonus ! I stuck the stripped end of some fresh new wire up from below, and soldered it in place -- the fresh drilled copper made a good clean surface to bond to.

When I was done, I put the top (flange) into the bottom (cup) and started squishing the seal shut. At first, the squishing has to be radially inwards, to reduce the diameter of the pried-open part. Then, the blows (I used a ball peen hammer on top of the vise) need to go downward, to crimp the lip shut against the flange. In practice, it's really sort of a rolling process. Anyway, I'm sure the joint isn't waterproof, but it did get pretty tight -- with quite strong pressure from each hand, I can't make it rotate. Just to be sure, I'll probably smear some goo on it.

What I love about this job is that it's one of those old-world things you could do on the side of the road, if you had to (but you don't...) One reason I got my 914 was to counterbalance my other car ('97 Saab 900 turbo) that is so friggin' complicated, that only a licensed dealer w/ the right computer can even start to mess with it.

Anyway, as your pictures show, there are two halves to the AAR : the electrical half (a simple heater) and the mechanical half (coiled up bi-metal spring that rotates a valve. After doing this job, I'd encourage anyone with AAR woes to open theirs up and try giving it a fix. The stuck valve can't be much more than you found - twist it back and forth a little, and maybe lube it.

The electrical half isn't so bad, either. The heater element has about 13 ohms of resistance, which dissipates 11 watts (P=v^2/r). Anything with around 13 ohms resistance will work, but it must be able to stand up to a steady state dissipation of 11 watts (plain electroniconic-store resistors are only 1/2 watt or so, typically). The wire that's in there is probably either tungsten or nichrome. If you can get your hands on either, cut a length that measure 13 ohms or so, and wrap it into a coil around a nail or something. An easier method might be to clip a section out of an old toaster -- you needed to go to thrift store anyway, you CSOB.

Anyway, I'm off for home (laundry, etc.) and I'll try to take a snap of my "new"
AAR and post it tomorrow.

Cheers,
Jeff

[Dave_Darling]

Jeff, that's exactly how I opened up my old AAR.

I thought the power was only ~1 watt? Hmm, you might need some really beefy resistors if it's actually 11 watts!! (IMG:http://www.914world.com/bbs2/html/emoticons/sad.gif)

13 ohms worth of resistors (that can deal with the current!!) will heat up just about as quickly as the stock "resistance wire" did--so you can swap in plain resistors instead of the fancy wire stuff if you need to.

--DD

[jsteele22]

Hi Dave,

Yeah, the power is around 11 watts (v^2/r = 12 * 12/13), which is a fair amount of power for normal resistors. But generating that much heat is the whole point -- it's kind of a mini toaster oven in there ! Ordinary resistors usually go up to like 1 watt, so a string in series/parallel would work but be kind of bulky; might not fit.

Anyway, to show off my handiwork (and my cat's butt) here's a couple of pictures.

[jsteele22]

and the other side, not butts about it ...