Atlas/Craftsman Serial Numbers and Bearing Dates (if applicable) For Database Entries

[wa5cab]

@jpack,

After thinking it over some more, I came to the conclusion that it is more likely that it is a 101.27440 rather than a 101.07403 with added QCGB.. It has the QCGB, the rectangular motor switch plate, and the selector handles are the late style.

I've assigned it serial number UNKNOWN000106, If you should ever, for some other reason, pull the spindle, please report the 4 dates engraved into the spindle bearings, or NONE if there aren't any. Atlas or Timken ceased dating the bearings circa 1953. The 101.27440 (and for that matter the 101.07403) was made up until mid-1957.

 

[Orangetruckman]

I am in awe of how awesome this site is!

I’m not 100% about this lathe other than it works. The mounting base has 3 holes and the belt cover is like 6” and 10” lathes. It has power cross-feed, the knob is off because it was broken by the PO. It’s a 12” model 101.07403, bed length 48”, Timken bearings (it’s tight, so I don’t need to pull the bearings. When I do, I’ll post the numbers.) and the stamping on the ways says “8622 S” I posted a pic. There’s no plate on the end of the ways. The picture in my profile is of the machine. I knew more about the machine than its PO did, so no good information on original date of purchase.
Here’s the details: I bought my Atlas lathe for $500 in North Plains, Or about November of 2020 from a guy, who’s grandfather owned it and passed it to his dad. The PO’s father had recently passed away and neither of the sons wanted it, so it was listed. He assumed that his grandfather may have been the original purchaser. It was used in a shop for repairing logging trucks for some time.
The lathe hadn’t been used in a few years and was covered with rat poo on the lathe and in the cabinet. The rust was just at the point of causing surface pitting, the leadscrew threads were pointy towards the headstock,
It came on an old school built bench with true 2x dimensional lumber and square drive hardware holding it together. The accessories included: 6” 3 jaw Wescott scroll chuck, the original 8” 4 jaw independent chuck, the open “C” style steady rest, 2 drill chucks, the original MT3-MT2 sleeve, lantern tool post, (these are the tooling for the lantern tool post) 2 LH, 2RH, 1 threading, 1 straight, 3 parting, 1 knurl, 1 boring bar, 3 dead centers, 4 different sized lathe dogs, 3 threaded 1 1/2” x 8tpi backing plates-sm, med, Lg, a drilling pad for the tailstock, a couple adjustable reamers with rust starting, almost a complete set of change gears, a MOLO with the threading section missing but I printed out the missing section. And finally a portable keyseater that I recently figured out that it’s not missing anything. Let me know if you have any questions.

 

[wa5cab]

Thanks. Is the nameplate with the model number on the rear or the right end of the bed? 8622 is the serial number. No one seems to know what the "S" stood for except that it was not Sears as it is also found on Atlas 10D and 10F of the same period.

If the lead screw is worn to a "V", it is worn out. A fairly common trick is to flip it end for end, cut the half inch part off of the left end and machine the right end down to 1/2" and thread the end. And relocate the right bearing to the left a few inches or use it on a 12x24 and find a lead screw off of a 12x36 and modify it to fit the 12x30. And replace the half nuts, which are still available from Clausing.

The MOLO's of the period had the threading section printed separately but not bound in. So it is a Version 3 by my numbering system. Technically, V2 and V3 are the same except for the loose threading section.

 

[Orangetruckman]

Thanks. Is the nameplate with the model number on the rear or the right end of the bed? 8622 is the serial number. No one seems to know what the "S" stood for except that it was not Sears as it is also found on Atlas 10D and 10F of the same period.

If the lead screw is worn to a "V", it is worn out. A fairly common trick is to flip it end for end, cut the half inch part off of the left end and machine the right end down to 1/2" and thread the end. And relocate the right bearing to the left a few inches or use it on a 12x24 and find a lead screw off of a 12x36 and modify it to fit the 12x30. And replace the half nuts, which are still available from Clausing.

The MOLO's of the period had the threading section printed separately but not bound in. So it is a Version 3 by my numbering system. Technically, V2 and V3 are the same except for the loose threading section.

No problem sir! The name plate is on the backside in the center. I’ve kinda wondered if the “S” in the serial # meant that it was seasoned and ready for use? Stamping a serial # at the same time help identify which ones were ready to be used would make sense?

On the lead screw, I measured thinking about swapping end for end, but the wear would still be about the same area, ugh. So, I bought new half nuts and acme rod with a keyway already milled in for like $200 + shipping from mymachineshop.net just gotta cut to length and turn to size on both ends.
Good to know about the manual. By chance, would you happen to have a guesstimate on what year my lathe is? Thank you in advance and doubly thank you for all the help you provide everyone!

 

[wa5cab]

My rough guess as to the date of manufacture is late 1943.

"S" for Seasoned is a possibility. However, all after about 101.07403 S/N 015000 don't have it. But the statement in the MOLO about the seasoning is in all versions from 1 through 9. And probably in Version 0 but I have never seen one (I had been keeping records of the MOLO's for 3 or 4 years before I discovered a catalog photo showing that a version earlier than V1 existed). Incidentally, the full model number of your machine should be 101.07403-2075, which pins down the bed length.

 

[Orangetruckman]

Thank you sir!

 

[mechmike]

I bought my Craftsman 12" lathe to make a back gear for my Atlas milling machine. I learned a lot. Can anyone help me find out the age of the lathe? Model #101.07403 plate on back of bed. Serial # stamped on top of bed right end 34155.
Thanks

 

[wa5cab]

OK. Late 1951 or early 1952. Should you happen to, for some other reason, need to pull the spindle and bearings, please report dates on both cups and cones. In the meantime, what is the bed length (42" or 54")? And was there anything else stamped on the front way before and/or after the serial number.

 

[robert574]

Welcome mechmike! Let me see...
27026 was the highest s/n for that model dated 5/10/1950 on the spindle bearings. If I graph the s/n's we have and draw a line it comes out to be early 1954. I'll try some other model numbers and see how they fit in the same time range.

Edit: This is bringing back memories of the last time I tried to do it. Had to throw out all the data points except for bearing dates and receipts. A LOT are calculated (who knows) or a guess based on what someone else guessed at. Ok ready, my best guess (based on 2 good dates) is Sept 29, 1952. I think that's the newest it could be. There are a few slightly older bearing dates mixed in, but I try to use the newest bearings and leave out any old stock. Seems right.

 

[wa5cab]

Robert,

Sorry but the highest serial number 3/8" bed 12" machine currently known to have been dated is a 101.27430 with serial number 035633 dated 1952/12/20. Or a 101.27430 with serial number 038396 dated 1951/09/10. But the latter example is well out of sequence.

"Calculated" in general means a straight-line calculation (which means assumed constant production rate) between the unknown date serial number and two other serial numbers with known dates. However, we know from the existing evidence that Atlas did not practice FIFO with their bearing stock. And we also know that the production rates fluctuated widely over the years. One reason we know the latter is that had Atlas been selling as many per month in 1957 as they had been in 1942, they probably would not have discontinued the 3/8" bed machines. And "Guess" means that I took a quick look at the list and just guessed a range of dates from two or three other known serial numbers and bearing dates, but I did not do any actual calculations. And we have several examples of earlier dates with higher serial numbers. Some are noted as being out of sequence. And some of the dates are from "family history" and the like. Most of these are noted in the date reference field but I inherited about a third of the serial numbers and did not initially have that field and some others. For example, the serial number and name tag location field was only added recently.

Also, do NOT include any 10" machines in 12" calculations/estimates/guesses or vice versa as all 10" (except for the QC 10" models made between late 1947 and sometime in 1951) appear to have had unique serial numbers whereas Sears started the serial numbers over with each model number change. So there was a serial number 1 of the 101.07403 but there was NOT a serial number 1 of the 10F. Actually, we do not know for certain what Atlas or Sears started each serial number pool over at. However, we do have a number of 3-digit serial numbers so if they didn't start with 1, the most least that they could have started with was 100.

Finally, any date fields in the serial number database beginning with a blank space are suspect for some reason or other. Only known/reported bearing dates begin in the first column.