History of my shop

[Geof]

I started this rather than hijacking Travis' thread on the cost of setting up a home shop.

This started out to describe what it cost to start my first shop but it also turned into a description of how I modified machines to make them more useful for my purposes and how my business developed over the past 29 years.

I started my shop, which at the time was a home shop in the basement of a large house, in 1981; it was a 'real' shop in that I intended to make a living with it. I am only mentioning major expenses but keep in mind the date when looking at the costs, and I am not including minor tooling.

The first machine I bought was a vertical bandsaw at around $500 then came a small mill with 18", 8", 8"+4" xyz travel and a 12" dia over the bed lathe with 36" between centers at a combined cost of about $6500. These were Tiawanese machines of moderate quality. A horizontal band saw came next at about $500 and a bench drill at about $300. I have all these purchases recorded in a journal I kept back then.

There were no other major machine purchases for about 5 years then I finally developed my own product which started to sell. This led to the acquisition in early 1987 of a used Tos R5 Capstan lathe for $5000 and a Herbert #7 Turret Lathe again used for $5500 toward the end of '87, along with a couple of bench mill/drills at a combined cost of about $2000. At about this time I stopped keeping the journal so I don't exactly remember all the small machine purchases. Also by this time I had moved out of the basement into proper commercial premises.

About a year later I picked up another Herbert #7 for $850; surprisingly enough this was in better shape than my first one but was incredibly dirty. By this time my product line was moving; it consisted of aluminum parts parted from 2" and 2-1/4" round bar which is why I needed big machines.

The parted-off parts got intersecting through holes 0.880" and 0.630" diameter which initally were drilled and bored on two bench mills; drill the part in a fixture on one mill then move the fixture to a second mill and bore the hole. Each mill was fitted with a base that the fixture locked into for alignment. Drilling and reaming was not practical because the first hole was not through the centerline of the bar diameter so the drill wandered several thou; similarly the second hole intersected the first and the drill wandered a lot making boring essential to get the hole to hole tolerance needed.

This two machine procedure worked but it was time consuming and a better solution came when I bought my biggest machine, a Cam Control Herbert #4 Auto which was 12 feet long, 5 feet high and 6 feet wide weighing 16,000 lbs for $2500. I stripped off the cross slides, original motor and gearbox, and chuck just leaving the turret and the spindle intact; in the process reducing the weight to 'only' 12,000 lbs give or take a bit. The turret had four tool stations and the turret slide was moved forward and back by a big helical drum driving a dog, indexing the turret once every stroke. Two of the tool stations had a cross-slide affair for mounting and adjusting boring bars while the other two were fixed; I put drills in the fixed stations and boring bars in the cross-slides. In place of the chuck I built a faceplate affair that could mount the fixtures I used for doing the parts on the bench mills; these fixtures were positioned so they rotated around the centerline of the hole. Now the holes were automatically drilled then bored at a single fixturing in fraction of the time needed for the bench mill operations. It was fun watching this machine run because I boosted the speed to 1800 rpm, about three times what it was designed for. The spindle was about 5" diameter and the bearings were huge; after a few hours at 1800 rpm they got far too hot to touch but this thing ran for more than 9 years doing tens of thousands of holes. I never bothered precisely balancing the fixtures that were used on the faceplate because the machine was so huge it didn't matter if they were several pounds off; sometimes however, depending on what fixture was mounted, the machine would hit a resonant vibration and make the floor bounce in wave patterns all over the shop. I also picked up a smaller Herbert 2D turret lathe for about $1300 and modified it similar to the big guy for doing smaller parts.

The two modified machines improved productivity so much that the hole boring process ceased to be the limiting step (until I got into CNC); but now parts could not be parted off quickly enough. Fortunately I got lucky, there was a machine shop liquidation auction and in early 1993 I picked up two Herbert #5 turret lathes and another 2D for a total cost of $7500. One of the #5 turret lathes had an electrically operated collet chuck and an electric barfeeder and could handle up to 2-5/8" bars 12 feet long; it was really something to see this machine turning and parting 2-1/4" 6262 barstock at 2000 rpm without being slowed down to open the collet and feed the bar. One of our parts has a hemisperical end and this machine could cut this hemisphere with a form tool in one go. Needless to say parting was no longer the rate limiting step; actually we had excess capacity at this point.

If you do a quick total my machine expenses so far were about $30,000 for a total of eight big machines plus my original small lathe and mill. Also by now I had picked up a couple more bench mills and a Millrite knee mill for making tooling and these brought the total outlay to about $38,000. There were other incidental expense like a few thousand for a compressor and a forklift but I think my final outlay pre-CNC was less than $50,000. Between 1993 and 1999 I did not purchase any additional machines and simply worked on upgrading fixturing and procedures to keep ahead of demand.

By early 1999 it started to become obvious that my way of doing things on manual machines with dedicated fixtures and tooling was starting to choke. Instead of changing setups on machines I had different machines set up for different operations; the operators moved from machine to machine for each operation to complete a part. The problem was once the capacity of the machines was saturated I could only increase output by buying and modifying more machines and the supply of cheap machines was drying up. For my $50,000 investment, plus a lot of my hours designing and making tooling and fixtures, I was doing pretty good; I had 4 employees and annual revenues of a bit less than a quarter of a million by 1998 but drastic changes were needed for further growth.

By mid 1999 my wife, who ran all the financial side of the business, and I decided the only way we could progress further was to get involved in CNC. For once we actually made a business plan and intended to analyse what operations could best be done by CNC machines and embark on an orderly transition. This intent lasted for about three weeks and then I saw a Haas HL1 for sale at the local Haas dealers Open House and snapped it up for $38,500. This led to a few weeks of hair tearing on my part as I taught myself G code on a lathe without the correct manual for the machine but I prevailed. Within a couple of months the HL1 was fitted with a hydraulic barfeed of my design and was popping out parts so fast we could not keep up with it so obviously a VMC was going to be needed.

To make the story short in April of 2000 I bought a new VF0 and between then and the end of 2008 bought another Haas machine on average about every 5 months; all our plans to do things orderly went right out the window and we just grew and grew. The third machine was a Haas HL1 Big Bore then two MiniMills, a SL10, a Super MiniMill, another SL10, a VF2, two more Super MiniMill, a GT20, another VF2, a GR510, another VF2, a TL1-P, a TL2-P, a TL1, three more Super MiniMills, a TM3-P and another GT20. The number of employees grew also but not as fast; now it varies between 8 and 10 in the manufacturing side and another 8 in the offices. Needless to say annual revenues more than kept pace with the growth in the number of machines and employees and I am still doing pretty good.

[Switcher]

I had 4 employees and annual revenues of a bit less than a quarter of a million by 1998 but drastic changes were needed for further growth.

Can you post a comparison, of how much the cnc vrs. manual setups made last year?

[Geof]

Originally Posted by Switcher Can you post a comparison, of how much the cnc vrs. manual setups made last year?

Ooh should I really do that, you might be able to figure out how much I make.

Based on my experience and very roughly speaking the productivity per employee involved in manufacturing is about 4 to 5 times higher for CNC than it was for manual. This comes from an approximate doubling of the number of employees and a tenfold increase in annual revenue.

I never kept really detailed records but there are some procedures and parts that are illustrative of the difference. The hole drilling/boring operations I mentioned was about 3 to 4 minutes per hole using the fixture on two bench mills and this went down to around 1 minute 15 seconds on the big modified machine; these times also include loading and unloading the machine. These holes, done individually, on a VMC take fairly close to a minute including the reload time, so there is not much improvement here. However, and this is an important however, on the VMC we use a rotary fixture that holds six or eight parts and allows both holes to be done at a single fixturing; this brings the per hole time down to something like 12 or 18 seconds. The improvement is not just CNC it is CNC coupled with creative fixturing.

[Switcher]

Thanks,

Great post you have started!

I agree 100% about fixtures. I ran mig welding robots back in the late 90's, that had twin cells, load side one while the robot welded side two, etc... It worked great, the robot never stopped, cycle times maybe 10min per side. One guy could run four cells (2 - bots) easy.

I bet you had a smile on your face, the first time you dropped those cycle times with the cnc, compared to the manual setup.

[Geof]

Originally Posted by Switcher ....I bet you had a smile on your face, the first time you dropped those cycle times with the cnc, compared to the manual setup.

I never saw the dramatic time saving in a single jump because it never happened in a single so the smile was spread out. Initially the hole boring was done 4 at a time in a pair of Kurt vises with custom jaws and as I mentioned this was not radically faster than our big monster could do. Then we moved to three Kurt Duolocks which now held 12 parts so the times got a little better. So the final step to the rotary fixture was so far removed from the manual procedures the overall reduction in hole boring time was not obvious.

However, one part did give rise to a satisfied smile, the one I show in Post #24 in this thread: http://www.cnczone.com/forums/showth...t=21953&page=2

We used to make this part on manual machines and I can say it was a challenge and slow at something like 48 minutes from start to finish. When we switched to CNC initially it was done in three setups, four at a time in vises and the time was brought down to something like 18 minutes; that led to a satisfied smile. Then I put in about 10 working days making the rotary fixture shown in that post, and reconfiguring the program to work on the parts in three orientations and that brought the time down to under 8 minutes per part which made the smile even more satisfied.

But it is not just the saving in time that is a benefit it is that my guys can run two or more machines with cycles times anywhere from 30 to 60 minutes. The fixtures are duplicated so one is reloaded while the other is running and then switched into the machine in a matter of 30 seconds. Sometimes the guy running two machines still has enough free time he can spend 10 or 15 minutes every hour doing Sudoku puzzles because he literally has nothing else to do.

[TravisR100]

Geof, thanks for the history. Enlightening.

[Kool Parts]

The improvement is not just CNC it is CNC coupled with creative fixturing.

This is the key to it all. Not a lot of shops where I am, but every time I see someone machining a 2+ operation part I immediately see they are wasting time in the fixturing.
Gary