AE D1 Continental clone? project

[unbeldi]

Poplar just notified me that I got it backwards, which makes sense because those patent numbers are actually for the A1 mounting.  I had commented about that at one time in another thread..... 

Here is that thread:  http://www.classicrotaryphones.com/forum/index.php?topic=10630.0

[unbeldi]

Zinc is almost 3x as heavy as aluminium. Specific mass is ~2.7 g/cm^3 for Al, and ~7.1 for Zn.

It would be nice to actually measure the mass of these housings to compare. I haven't had two of them completely stripped of paint and other parts.

[unbeldi]

Just looking at your 'inside' view of the phone again, and it appears to me that the surface doesn't really look like Al, rather has the appearance of an alloy.

[TelePlay]

Just looking at your 'inside' view of the phone again, and it appears to me that the surface doesn't really look like Al, rather has the appearance of an alloy.

Good eye! I do have an aluminum D1 base for comparison. That base is stamped III 35 inside in vermillion. Both have "D1" and "USA" stamped in the back under the cradle mount and neither have the patent info at the bottom of the back base.

Both have been stripped of paint and are empty of parts. First thing I noticed is the cradle mount in the II 37 "project" phone is slotted all the way through and seems to be a "post like" casting versus the solid "shelf" in the '35 base and only slotted in about a half inch from the top.

I also noticed the '37 base has K-7 stamped into the metal near the II 37. The '35 base has a 5 point raised star and to the right of it A-5 stamped into the metal near the III 37 III 35.

The '35 base has a mass of 174 grams. The '37 base is a whopping 400 grams, more than twice the mass of what seems to be a true aluminum base. When I get some time, I'll calculate the volume of each base and that should give a close approximation the density of each metal used.

So, poplar1, do we need another topic to track D1 bases as to year, stamped marks inside and exterior markings? Is there something worth tracking here? Your call on that.

The photos follow in sufficient resolution to see the detail. The '35 aluminum base is always on the right side. The exterior of the '35 has been polished; the '37 as is right after paint stripping. I also noticed the '37 base, which seems to be something other than aluminum, has black or dark spots within the metal which I would guess is something not getting mixed quite well enough before the alloy was poured. The '35 do not have such spots.

The surface of the inside has not been touched in any way - they are as received. When I get my digital microscope out, I'll take pictures of the internal metal stampings.

[unbeldi]

Good stuff !

I have to check the variability in density of aluminium and zinc alloys, but I didn't think it would account for the low ratio that you observed  (400 / 174 = 2.3). Clearly the volumes are different for the two types, and if you can measure that it would be superb. 

We have to look for sources that specify just what WECo used for alloying zinc, but zinc die casting using pressurized molds was developed in the 1920s. These alloys contained a small amount (3-4%) of Al, magnesium, and copper. Because of the composition they were called ZAMAK, after the German names of the metals.  It wasn't until the 70s when Zn-Al alloys with high aluminium content (>40%) became common that only required gravity molding.

There is a series of publications by WECo that presents the materials used from around the world.


PS: It seems that Zn alloys with about 4% Al have a density of 6.6 g/cm³.  So the ratio of Al/Zn-alloy is down to 6.6/2.7 = 2.44, not too far away from 2.3, which could easily be explained by change in volume between the two types.

[unbeldi]

I didn't realize that the cord hole in the back of the body was what appears to be a brass insert.  They obviously discovered that a raw metal edge would probably, over time, tear into the cotton or silk finish of the cords.  So it was easier to insert a grommet, rather than polish inside the hole.

[unbeldi]

Good eye! I do have an aluminum D1 base for comparison. That base is stamped III 35 inside in vermillion. Both have "D1" and "USA" stamped in the back under the cradle mount and neither have the patent info at the bottom of the back base.

So you have a nice first-hand example of Poplar's distinction of the 1935 transition from Al to Zn, by which the last Al bodies do not have the patent markings.

[TelePlay]

I didn't realize that the cord hole in the back of the body was what appears to be a brass insert.

Yes, I've stripped 4 D1s now and they all had the brass "grommet."

Now, as for density, I used a small plastic container ( just large enough to take the base), placed it on my scale (covered with Saran Wrap to keep it dry), put the scale on a rack to keep water from getting into it from the bottom and filled the container with water until the meniscus broke and water ran out from over the top. Once that stopped, I put the base in the water and let the water drain over the top until it stopped. I then carefully removed the base. I had set the tare on the scale to "0" before putting the base into the water so when the base was removed, the scale reading was the amount of water that ran out due to the volume of the base. I use the grams option on the scale.

The '35 base displaced 74 grams of water; the '37 base displaced 68 grams. I guess that had to do with the post on the '37 versus the ledge or shelf of the '35 (to hold the handset cradle). The notch in the top of the '35 is really only about a 1/4" deep, just deep enough to match the lug on the cradle).

A gram of water is 1 cubic centimeter so diving 174 by 74 and 400 by 68 gave about 2.35 g/cm³ for the '35 Aluminum base and 5.88 g/cm³ for the '37 alloy base. Now, I suspect that base is not pure Aluminum and the other base is not pure Zinc. But, this is what I got and I did the experiment for each twice, using the averaged numbers.

I wish a had a large graduated cylinder to determine the volume of each base. A plastic container and dealing with the water meniscus is not an ideal way to determine volume but it's close. If I can think of a better way to measure volume, I'll do this again.

EDIT: I didn't realize this topic existed until poplar1 sent me the link to it, this link.

http://www.classicrotaryphones.com/forum/index.php?topic=11422.msg125730#msg125730

[unbeldi]

A gram of water is 1 cubic centimeter so diving 174 by 74 and 400 by 68 gave about 2.35 g/cm³ for the '35 Aluminum base and 5.88 g/cm³ for the '37 alloy base. Now, I suspect that base is not pure Aluminum and the other base is not pure Zinc. But, this is what I got and I did the experiment for each twice, using the averaged numbers.

Can't argue with results of an experiment.

However, both numbers do seem low.  I can't imagine an Al alloy being substantially (not even 5%) lower in density than the pure bulk metal, it being so light as it is. Its liquid density is 2.4 and the result of any alloy should at least be that.
Almost any alloying might in fact increase it, as the crystal structure wouldn't change dramatically.

For Zn the liquid density is about 6.6 vs. 7.1 g/cm³ for the solid, and the same applies for low content alloying with aluminum. I was going to point out before that the alloy has about the density of the pure liquid, which might be considered the lower limit.

[unbeldi]

EDIT: I didn't realize this topic existed until poplar1 sent me the link to it, this link.

http://www.classicrotaryphones.com/forum/index.php?topic=11422.msg125730#msg125730

Yes, David has been very diligent in his efforts to record that stuff, a great resource. We just have to figure out how to optimize the display of such compilations given the limitations of forum content formatting, and the shear labor involved in updating the data.

[unbeldi]

PS: It seems that Zn alloys with about 4% Al have a density of 6.6 g/cm³.  So the ratio of Al/Zn-alloy is down to 6.6/2.7 = 2.44, not too far away from 2.3, which could easily be explained by change in volume between the two types.

Indeed, now that you got a difference value of the volumina of the two bodies, we can judge this assumption.  74/68 ~ 8.8 %.  Even if your results have a systemic error of being too low, it may be argued that the difference is more accurate, canceling the error.   So, 2.3 + 8.8% = 2.5, which is a bit closer to 2.44, albeit higher.  I have no problem with the error margins in all of this.  It seems somewhat consistent, qualitatively.

Hmm, or did I get this backward?   Should it be   2.44 - 8.8% ==> 2.24 .... even closer.

[TelePlay]

Yes, I also noticed that the numbers were low. That's why I want to find a better or more accurate way to measure. To get 2.7 g/cm³, there would have to be additional metal volume of about 9.5 ml or 9.5 cm³. To get 7.1 g/cm³, that would take about 11.5 cm³ more volume.

That's a lot but watching the meniscus issue when doing the experiment makes me want to find a better vessel to make the measurement.

[unbeldi]

Yes, I also noticed that the numbers were low. That's why I want to find a better or more accurate way to measure. To get 2.7 g/cm³, there would have to be additional metal volume of about 9.5 ml or 9.5 cm³. To get 7.1 g/cm³, that would take about 11.5 cm³ more volume.

That's a lot but watching the meniscus issue when doing the experiment makes me want to find a better vessel to make the measurement.

To eliminate the meniscus problem, you might try adding a few drops of dishwashing soap to the water.

[TelePlay]

To eliminate the meniscus problem, you might try adding a few drops of dishwashing soap to the water.

That helped. I also found an old glass coffee carafe that had a pour lip that the base would just fit into. Reset the experiment so the pour lip would drop expelled water away from the scale and this time waited until the water stopped flowing from the pour lip. Having a small point run off is better than the complete top run over that I used in the first measurement.

For the Aluminum base, I got 68, 70, 70 and 68 grams in four tests = average of 69 grams. 174/69 = 2.52 g/cm³

For the Zinc allow, I got 62, 64, 66 and 64 grams = average of 64 grams. 400/64 = 6.25g/cm³

Both numbers are closer to but still below 2.7 and 7.1 g/cm³

I'm stumped right now. Can't think of a better way to do it. Might have to sleep on it.

If I knew someone with equipment to do an qualitative and quantitative energy dispersive analysis of the metals, I'd like to see what they are made of, but, I haven't had any contact with anyone in that kind of a lab for years and I'm not about to pay the money to have it done in a commercial lab.

Scrap yards have that sort of equipment to determine the content of metals being brought in for scrap, don't they?

EDIT: I might try creating a displacement can and measure the water by volume rather than weight. The amount of water displaced is only about 1/3 of a cup - that's all the metal that was needed to make a D1 base.

[unbeldi]

That's a big improvement. Nice work.

I expect the actual density of the Zn alloy body to be around 6.6 g/cm³ and have around 4% Al as major addition to Zn. This was pretty standard for decades. For the Al body, it can't be that much different from the pure metal.

Given your household equipment, an error of 5% seems totally within possibility. Did you, for example, make sure no air bubbles remained in the screw hole posts to which the base plates would be attached?

You have two measurements involved, volume and mass, and when combined in a mathematical operation, the error propagation increases the deviation from each individual margin.  Your volume measurement had an error of +/- 2 cm³ (for Zn) which is a range of 6% alone.

I doubt scrap yards do much of chemical analysis, unless they are specialties recyclers. Commonly used alloys such as these are pretty much standardized anyways. But WECo did have their own smelter facilities for metal recycling, but I doubt they made their own specific alloys. By 1935 this process was well established.