errors in The Ultimate Tone by K. O'Connor

[bob p]

Europe: You come across a nice WW-II era German military radio transmitter, yet you discover that one of the panel meters is missing a small screw. Eyeballing the size you grab an M3 screw from the parts bin, screw it in and continue on your way.

US: You come across a nice WW-II era US military radio transmitter, yet you discover that one of the panel meters is missing a small screw. You are now ready to kill yourself...

Kill myself? What for? During WW2 Motorola used fasteners that were of ASME/SAE standardized sizes in the USA. These standards were established as early as 1912-1916 and remained in effect throughout the war. These fasteners are still abundant over here. If I need a 00, 0000 or a 0000 gauge screw, I just take one out of the parts bin, just like you'd take an M3 out of your parts bin.

I think you'd have just as much difficulty finding a proper thread to fit a British or Canadian product from that era. The Allies had a lot of headaches related to lack of interchangeability of parts during WW2 because there were different screw thread standards in place in the USA, Canada and the UK. This was addressed in 1949 with the adoption of the Unified Thread Standard.

I think the take home lesson is that it's easiest to find parts for old radios in countries where they originated and where supply lines existed because they were widely used. During the WW2 era Denmark was either a Nazi protectorate or a Nazi occupied territory. If Denmark, Germany and Europe in general had all adopted the same metric thread standard, then an M3 screw replacement is no big deal. A Motorola radio, on the other hand, is a foreign object built to another set of standards, and it's reasonable to expect that you might have trouble finding ASME/SAE standard parts where you live. But where I live they're very easy to find, and most people that work on old radio gear around here either have these fasteners on-hand in their bins or they have ready access to them. When the Unified Thread Standard was adopted in 1949 it eliminated the 0000 through 000 sizes, but there are still lots of those parts available in their country of origin.

As you might expect, I have a hard time finding the tiny Japanese metric screws for repairing Nikon gear. I'm sure that they're easy to find in Japan.

[bob p]

Personally, I find much easier to convert in either Standard or very limited metric. Standard measures, I only have a few to worry about - inches, yard, miles. Those three units pretty much cover anything I will use. Also, in metric specifically caps, I find it easier to work with microfarads and picofarads. Even though technically incorrect, trying to convert between about ten different potential prefixes to choose from gets a little mushy. I am slowly learning to convert .047uF into 47 nF, but if we all got into milli, centi, deci, deca, etc. for caps, I'd go bonkers.

Would I be correct in assuming that you do the amp thing recreationally, and that you don't have a formal background in science or engineering?

The reason that I ask this is because dimensional analysis is a foundation skill for the science/engineering crowd. It's just not possible to do all of the math that goes into science & engineering without it, and once you have that skill mastered then all of the unit changes become trivial computations.

4n7 works for me. But then I have a hard science background so the SI abbreviations are second nature to me. I greatly prefer to read schematics that don't have leading zeroes, decimal points, and padded zeroes in the parts values. Typically the schematics that use these sorts of antiquated notations are antiquated schematics, and it's almost never possible for me to obtain an original schematic, or even high resolution copy of the original.

Vox schematics have to be the worst. The only original schematics that I have access to are 20-30 year old scans that were made at low bit depth resolution from low-contrast photocopies of the originals. The lines are faded and the zeroes and decimal points are hard to read. The tweed Fenders have the same problems. They're more contrasty than the Vox schematics, but they're all at very low bit resolution which makes the lines jagged and the numbers hard to read. Does that 5E6-A schematic call for a 0.00025 uF cap going to the treble pot? The layout diagram isn't clear ... does it say 25mmF? It's hard to tell on an old low-res scan, and it's frustrating when the layout diagram and the schematic for the same circuit use two different naming conventions.

Sometimes these values become more and more illegible as schematics go through several generations of photocopying before they finally get to scanning. Sometimes scanning is done to a lossy format like JPEG, which introduces blur into the image. Sometimes the image gets mangled a lot before it finally makes it's way to a non-lossy format. Hopefully the final scanning is done well, so that the information will never again be lost during replication. Unfortunately there were lots of schems that were scanned long ago, at low resolution, and the only way to get clear schematics is to re-scan them or to re-draw them, which adds the risk of more transcriptional errors.

Leading/trailing zeroes and decimal points are a serious problem in the context of duplication. During photocopying or transcription the leading zeroes can become ambiguous and decimal points can disappear, or blend in with background noise. 25nF is a lot easier to read, it takes up less space, and maybe it's less likely to get mangled when the schematic gets recopied. It's probably worth the effort to familiarize yourself with all of the other standardized units of measure because you're guaranteed to see them in the future.

[Gnobuddy]

Stop complaining y'all, or I will force the lot you to go back to using cm instead of Farad

Apparently I missed out on cm - it means "centimetres" to me!

US: You come across a nice WW-II era US military radio transmitter, yet you discover that one of the panel meters is missing a small screw. You are now ready to kill yourself...

In college (then newly arrived in the USA) I once had to design and construct a simple water-cooling system for an optical crystal (part of a tabletop laser). I measured the diameter of the cold-water pipe in the lab with a pair of vernier calipers - and was a bit confused, as it measured 0.98 inches. So I shut off the water, twisted the cap off the end of the pipe, and took it to the hardware store so I could find the right parts.

The hardware store guy takes a quick glance and says "That's a half-inch water pipe."

I say, "It can't be, it measures almost one inch in diameter."

Hardware store guy grabs a threaded nipple off the shelf, marked 1/2", and shows me that it fits the cap I brought with me. Okay, he's definitely right. But what on earth?

Back at the lab, I measure the inner diameter of the pipe. It turns out to be about 0.8". So a half-inch water pipe has an inner diameter that's not half an inch, and an outer diameter that's not half an inch.

This was before the Internet. Much later, I track down an experienced machinist, and he unravels the mystery. Decades ago, a half-inch water pipe had an inner diameter of half an inch. The outer diameter was nearly an inch, and that was standardized as the appropriate thread size for a half-inch pipe. Over the years, pipe manufacturing technology and materials improved, and the pipe walls got thinner, so the inner bore got bigger and bigger. And that's how you end up with a "half inch pipe" on which nothing actually measures half an inch!

"Metric" tire sizes are even crazier. A size of "P225/55R18" means a tire width of 225 millimetres, a dimensionless aspect ratio (treadwall height/tire width) of 55%, and an inner diameter of 18 inches. One unit manages to incorporate a schizophrenic mixture of millimetres, inches, and a dimensionless ratio!

Thoriated Tungsten's desire to kill himself in response to such travesties is understandable. I have a similar impulse. At the very least, a strong urge to beat my head against a convenient wall.

-Gnobuddy

[g1]

My standard arguement when trying to convince someone of the benefits of metric is usually "imagine if money didn't use factors of ten".
That usually does the trick, except with the tuppence and shillings crowd.

[Justin Thomas]

Hey Bob,

I enjoy the "hard" sciences conceptually, but I never could really grasp the math in the way it was taught. So yeah, this is all for fun to me. Social sciences I enjoy too, but not enough to dig in deep... as far as figuring things out, I'm adaptable enough to figure out the particular scheme used in a particular schematic and can convert it amongst itself. Consistency within the document...

Of course, most of the issue with cap values is with how the values are marked on the parts themselves. I have a funny feeling that Mallory knows most of their caps go into Fenders, so they use labeling consistent with Fender schema & payouts. The Orange Drops use a much more common labeling, and with some practice, I can read those too now. I also have some old pulled Mustard caps from a 50s projector that are all labeled in picofarads. But, the ranges of values we (as hobbyists) is fairly limited, so it is easy to memorize what a 10,000pF cap is in a different unit. (Is that .01uF & 10nF?
I could totally be showing my ignorance there!) The other thing is obsolete color codes. I have some old Bumblebee caps, all measure close to 22 nF on my meter. But of all the color codes I could find, there is no way in hell I can get that value by using the "convention" of the time. I've tried every color code I could find from the 30s to now, and it just don't make no sense. And no matter what order I read the 5 bands in, the value indicated makes NO sense - the indicated values are WAY beyond weird for the audio gear I pulled them from...

I've had some of the resolution problems you discuss, and it's why I keep my Jack Farr hard copy & GT Tube Amp Book print copies. I will pick up just about any old physical copy of a "golden era" text book I can find... I've managed to find good collections of the Tweed, Brown/White, & Blackface schema online. As far as the SF & later amps go, yeah, total crapshoot.

Agree on the Voxes - there are a couple good scans of the AC10 & AC15, but most of the others I find indecipherable... sometimes I just have to use a little bit of logic and just guess, and hope it's right...

Justin

[Justin Thomas]

Yeah, well, funny coming from the land of Toonies & Loonies... And stop Spocking your fivers! Actually, I still have some Canadian money in my wallet... And we still have quarters, nickels, pennies... none of that is tens...

Try British. After 2 weeks in Belfast, I was able to differentiate any denomination of coin in my pocket without looking, but only if I had variety of others to compare to...

Justin

[g1]

Haven't seen a Spocked fin for awhile now.
Different colour denominations is nice, we get a bit dizzy trying to cope with the all green stuff you have.

[Gnobuddy]

And we still have quarters, nickels, pennies... none of that is tens...

I don't think any country has money that only goes in steps of ten. It would take a lot of coins to come up with 98 cents in change if you only had pennies, 10 cents, and 1 dollar denominations to work with!

I think choosing currency values is a somewhat different mathematical problem, namely, what is the smallest number of denominations that makes it easy to make change without using too many coins? It's like old analog meters with their 1-2-5-10 ranges. A long time ago, a friend from India told me that his country did at one time have 1, 2, 5, and 10 paise coins, now long de-monetized.

I am curious, though, why in the USA we have $20 and $2 bills, but not 20c or 2c coins; why we have 25c coins, but not $25 bills; and so on. Canada has similar inconsistencies, too.

-Gnobuddy

[MarkusBass]

A long time ago, a friend from India told me that his country did at one time have 1, 2, 5, and 10 paise coins, now long de-monetized.
I am curious, though, why in the USA we have $20 and $2 bills, but not 20c or 2c coins; why we have 25c coins, but not $25 bills; and so on. Canada has similar inconsistencies, too.

As you mentioned, that "good" denomination of coins is a result of a simple mathematical problem. Please be informed that Polish currency (ZLOTY) follows the same rule (1-2-5-10-20-50). Exactly the same is with EURO. The coins are issued in €2, €1, 50c, 20c, 10c, 5c, 2c, and 1c denominations. I thought that most countries in the world follow the same rules. It seems to me that USA and Canada do not follow these rules because of historical reasons (monetary system "copied" from Great Britain?).

Mark

[MarkusBass]

Stop complaining y'all, or I will force the lot you to go back to using cm instead of Farad as the unit of measure for capacitance. One pF is about - yet not quite - 0.9 cm. (Or maybe it was the other way round...? )

This is rather impossible. Can you give a source of such information? It would mean that capacitance can be measured (in the US) in cubic root of teaspoons .

Mark

[Justin Thomas]

I'd caution everyone to never put ANYthing beyond good old Muhrikan "Redneck Ingenuity."

Justin

[Dave H]

The resistor/capacitor number code can be confusing. It's not so bad when it says 103 on a resistor which is 10k ie 10 x 10^3 but when it says 100 at first glance you could be fooled in to thinking it was 100 ohms but it's not its actually 10 ohms ie 10 x 10^0 and a resistor marked 1200 is 120 ohms not 1k2

[Jazz P Bass]

And then 'someone' decides to Not apply Any markings on surface mount capacitors.

Gee.
That's helpful.

[Tom Phillips]

... Social sciences I enjoy too, but not enough to dig in deep...

"Social science" Now there is a phrase that is far more confusing to me than the unit soup portion of this thread. I wonder who the weenie was that appropriated the word "science" for such a use.

[Justin Thomas]

I think the word "science" can be applied, as in, the methods can still be "scientific" in nature - hypothesis, experiment, conclusion, repeat, etc. Except they're more likely to use Statistics than Differential Equations. I mean, I don't think EVERY science requires the use of math-beyond-all-but-a-few.

I think Sociology is a science. Now, if I took a look at what category some degree programs fall into, we may agree... I guess that reminds me, I do have an AS degree in General Studies... Not sure I'd call it that, based on only ever taking ONE "hard science" the whole time, but that's what the Comm College calls it, so...

Justin