PM1440TL VS PM1440GT

[B2]

Hi Mark,

Thanks. Yes, I agree I need, and would love to have, more time on my lathe 1440GT. For those of us who are really hobby types or use the shop to make things for other projects... and so do not make a living in their machine shop this lack of experience is all ways an issue. The real reason I have a decent shop is that I ran into projects that needed parts that I could not easily get. I have done a fair amount of precision optics and was building a very specialized microscope and needed to be able to do very fine threading at high TPI for some of my optical mounts and slides. My old SB just could not do everything I needed any more.

I have had my solid state VFD installation running in one form or another for over a year now, and in the current version for most of the year, and have not observed any problems or failures. It works a lot better than the operator! I am still using the original motor as it seems to work fine. At least... good enough for me. I typically run it at 20-25Hz and certainly has no problem going faster, but when I want, it seems to run fine right down to 1Hz and I have not noticed any overheating or other issues. Maybe I need to run it much much longer? It even starts from a dead stop at 1Hz. I jog at 6 Hz. I doubt that it has much horse power at these low speeds, but it at least starts. As the U-tube video showed, the fellow had problems during acceleration and I think he even mentioned interacting with you. Yes, he said on his post that he switched to the Marathon E470 motor to fix his problems, but I don't think he mentioned you're getting it for him at 1/2 price! If you find another at that price let me know. I only see the ~$600+ adds.

Due to our previous discussion, regarding the proximity sense distance that one really needs for different cutting conditions, I went ahead and added a counter to my VFD design. I was going to do this anyway for my winding projects. Using this feature was how I obtained the TPI values for the non-factory gear positions. (see the table above). However, in this counter install, at the flip of a switch, it is also set up to start counting only after the proximity sensor trips. This way I get an accurate measurement of the fractional (1/10 turn resolution) turns of the spindle during the electronic braking process to stop. Also, who is to say that the Hitachi really breaks the spindle in the spec time? As I noted I have a 10 magnet Hall sensor set up so can resolve 1/10 spindle turn. I will get around to these measurements one of these days. Anyway, when running at ~25 Hz I noted a small amount of RPM fluctuations in the spindle RPM and so it occurred to me that I could probably measure spindle speed stability for various settings, loads, gears, frequencies, etc. I also have a spectrum analyzer in my shop that I could connect to the Hall sensor signal and do a bit more detailed analysis. It is also possible that this fluctuation is either due to my 10 magnets not all being perfectly space or that the counter is just not that stable. However, I did test one like it with a decent function generator before and did not see this behavior.

Stability perspective: At 60Hz the measured RPM fluctuations appear to be a few parts in 1000, say 1000 +-5. At 25Hz it is more like 1000+-10. At less than 5 Hz it appears to be more like 1000 +-50. Around 2-3 Hz I can hear the spindle speed gears clicking slightly which will stop if you put a drag load on the part. At 1Hz there is no clicking from the speed gears. Overall, for my work I consider all of these to be suitable.

At this point I cannot see any reason to change from my original stock motor as things seem to work fine. No acceleration problems and I have never really observed any issues once the machine is up to speed. However, all of this is TBD with more time and experience.

Dave L.

 

[Beckerkumm]

This is all interesting as I like the vfd conversion Mark did for my lathe. The Smart Brown has a 4-1 reduction using a Matrix clutch and an 8-1 backgear which is seldom used. With just the clutch I seldom need to run less than 30 hz but routinely run 90-100 hz . I like vector duty motors enough that if I find a Black Max, AB CM202, or Baldor IDNM NOS for less than $400 delivered I buy it for future machines.

There is a short write up over at Ruemema.de about the various spindle power choices by the guy who does maybe the most precise lathe restorations on the planet. He has hesitations on vfd conversions but I think mainly on lathes with no clutch or speed range choices. He talks about the torque at the tool moving when the hz gets too low but it is all above my pay grade to understand. Dave

 

[B2]

From my readings on HM it appears that a lot of folks are happy with Mark's VFD conversions. I think it is great that he provides this help to folks. I did my own because that is what I do... and because I wanted to get everything in the original PM1440GT stand cabinet with a minimum of wire runs to other boxes. It was shortly after I started looking into what was required that I realize that one really does not need the large transformer nor relays that are in the original machine at all. That is also why I made my own braking resistor so that it would fit in the same volume with the VFD etc. No big wires to run any where outside of the original electronics hole/cabinet in the lathe stand.

Dave L.

 

[mksj]

The 1440Gt builds I do for others, either are all self contained within the lathe or the VFD is mounted in a separate cabinet. It depends on the space one has, each build is tailored to the individuals specs. The VFD is mounted vertically in the headstock cabinet for proper cooling, they are not designed for horizontal orientation. The braking resistor is a aluminum cast unit, typically 50 Ohms/500W mounted to the interior bulkhead, there is no exposed connections to possibly come in contact with. Everyone does it differently, it is whatever works for each person. The control circuit topology is specific to the VFD used. The design and integration is dependent on a number of factors. I also integrated the complete control system into my lathe.

PM 1440GT complete VFD control system built into the headstock cabinet. MachTach tachometer and speed pot are in a separate enclosure mounted below the DRO.

Marathon E470 Motor installation, requires a new motor mount and modification to the belt cover.

ERL control system fully integrated into the stock cabinet, system is setup to accept single or 3 phase power.

 

[B2]

Hi Mark @mksj ,

Just to be clear, I am not trying to compete with you. I think the service and assistance you are providing to folks is great and I am glad YOU are doing it! It is clearly needed and I thank you for your contributions and comments. I used them in selecting the VFD and setting up my VFD parameters. While I am very happy to, and like helping folks, I have no interest in providing this sort of service other than maybe just to learn from the process, or especially to help others learn. I certainly have no interest in trying to make a living or earn extra money doing so. Time is far more important. Learning, teaching, and helping are my guiding objectives and I have lots of other topics I want to pursue before my time runs out!

Yes, "Everyone does it differently, it is whatever works for each person." This diversity is good as much can be learned from the process. I suppose the approach chosen depends a lot on one's design philosophies. I had objectives in doing my own VFD design. I wanted everything to be in the back of the lathe stand and I wanted a minimum of long or complicated wiring runs. Anyway as you stated: "what ever works for each person." It seems that my system works very well and there are no heating problems at all. Most importantly I got to learn a lot in doing my own conversion. You might want to try a solid state design sometime even if you want to mount the VFD vertically and put everything in an external enclosure. I would even be willing to help you do so if needed. I think you would find it takes less space, less time to build, and would cost less for the components. This would be especially true if the control electronics circuit board(s) were a pcb. You could even get it externally populated to avoid the point to point soldering that I did. Of course I was learning about VFDs etc. during the process, but I did spend considerable time designing, building, and figuring out how to mechanically interface the transistor circuit board to the cables. I posted my design in great detail so that others would not have to spend so much time at it. https://www.hobby-machinist.com/thr...tronic-components-pm1440gt-vfd-3-phase.95058/

I did not want to put the VFD and/or braking resistor under the non-enclosed head stock. I noted your design and thought it was clever, but difficult to get at physically. Just as importantly, I did not want either a big "Sea Slug" shaped braking resistor but, especially, the vented case VFD to be exposed to the dirt, belt debris, cutting debris, and especially the dripping oil from the over head gears and belts as well as the dirt and dust from the floor below. Yes, I know one could, and should build another shield/cover, for them when put in this location, but.....but why put them there if you do not have to do so. Also, I hated the thought of running power and, especially, control wires back and forth between the circuits etc. in the rear stand enclosure and the VFD in the non-enclosed stand under the head stock .... as well as to the front control panel. I learned long ago in electronics the fewer connections and the shorter the wires the better. Exposed wires which are subject to wear, especially control wires to the front panel are an additional potential problem. So, I did not want to hang an external enclosure on the wall.

WRT to the VFD mounting I suspect that Hitachi suggests vertical mounting for two reasons. 1) Visibility and access to the VFD keypad in a standard front facing cabinet. 2) Airflow cooling ... as "warm air rises." However, if one blocks the air flow from either below or above the VFD by mounting other things, i.e. cabinet walls or especially by heat generating devices such as relays, transformers, etc., this "warm air rises" conduction advantage of the VFD vertial mounting is eliminated ... or can make things worse. Air flow around any heat generators or heat sinks is important for their own life time... so non-crowding of these devices is important. Meanwhile, the built in VFD head sink fan provides greater air flow than, any "rising warm air" could possibly provide. One might worry about the fan failure, but these are rare and I suspect that Hitachi chose a decent small fan. So I mounted/oriented my VFD so that one can glance down behind the lathe, look through the Plexiglass cover plate to quickly see the display, which I find to be very handy. If one is willing to make seeing the VFD display a bit harder then it could be mounted vertially. Space is not the issue in the rear enclosure. Then I provided additional air flow, via the opening at the bottom of the cover plate (to be screen filtered for floor dust), via the fan opening in the enclosure opening. I also reversed the fan in the VFD heat sink to agree with this air flow pattern design. This air flow pattern also provides air flow over and through my open e-brake resistor assembly. So the air flow layout is consistent. I.e. air flows from the bottom left entry across the VFD heat sink, over the e-brake resistor bank and then to the cover plate exit. However, the best I can tell only having a vent would be more than sufficient. The fan I mounted in my cover plate does not seem to be needed, but is a VFD longer life safety factor. Furthermore I have checked things regularly and nothing ever seems to be more than barely warm. In addition, the VFD heat sink is mounted to a smallist Al plate at the VFD back which helps to trap the VFD fan generated air flow inside the VFD heat sink......... so heat from from the VDF heat sink flows into to the VFD back/bottom plate. If I had found heat to be an issue at all I would have had to do is also mounted this plate to the large enclosure backing plate via bar bracket (like I did to support the ebrake resistor bank) that would then be mounted to the large plate that is used for all of the electronic components in the enclosure. Then a thermal contact based conduction path would have been VFD heat sink to Al backing plate to large enclosure plate. But it has not been an issue at all so just did not bother with this last connection.

While I have not wasted time to cut open a "Sea Slug" shaped resistor, I suspect that they are constructed in the same manner as the power resistors that I used in my e-brake assembly.... a high resistance wire coil surrounded/potted in a somewhat thermally conductive ceramic material to insulate it from the Al cover. This resistance wire, when bare, can run at glowing hot without burning out, but hopefully it does not have to. The resistors are rated at a continuous usage wattage, but by the very design can dissipate far higher values for short periods of time. It is the time average power that is important. So when power is initially applied to one of these, it is cold and starts to warm up, but the heat diffusion to the Al surface is pretty slow even on lathe usage time scales. (If you have an electric stove top think how long it takes for one of the burners to get hot. These are also made with a resistance wire buried in a ceramic.) The lathe e-braking time, at least for non-monster sized lathes, is short compared to these thermal diffusion times. Likewise, for lathes the time between e-braking events is long compared to the e-braking time (but is user dependent). So there is a long averaging time for the heat dissipation. That is, this duty cycle is very low. So while I built an 8 resistor, 400 Watt e-brake resistor assembly, I doubt that one needs any where near this dissipation capacity for the lathe application. It seems to be way over kill. Since most folks have never worked in this field they have little appreciation for heat and power dissipation so it is almost always over specified. At one point in my life I designed power supplies so I did worry about, and studied and modeled, this quite a bit. (An application where this sort of e-brake power dissipation might be needed is one where a continuous control system was being used and so the duty cycle on the e-braking could much higher. An example of such might be in in the newspaper web handling process where very large heavy rolls of paper are constantly being accelerated and decelerate to maintain flow and to prevent delicate web (paper) breakage. As I mentioned, I started and stopped (e-braked) my lathe a bunch of times as quickly as I could and then touched one of my resistors to find it had only slightly warmed. It certainly was not hot to the touch. (This could be much different if I was trying to over come the inertia a 4' diameter x 3' wide paper roll!) This was without the enclosure cover fan in place. Even if they were to get hot they are mounted on a highly thermal conducting Al plate which in turn is mounted on the large Al plate that holds all of the other components. So heat in the resistors, if not air cooled, would travel to the resistor mount plate and then from this plate to the large Al plate. But it does not appear that this heat diffusion path is even needed.

Sorry for the Sea Slug (sea cucumber) analogy, but I sometimes scuba dive at Grand Cayman Island and every time I see one of these big e-brake resistors this is what I am reminded of...not very pretty and very slow moving.... https://en.wikipedia.org/wiki/Sea_slug#/media/File:Fish3923_-_Flickr_-_NOAA_Photo_Library.jpg . PS. There are some very pretty and smaller Sea Slug species, but they are less common around Grand Cayman.

Regards,
Dave

 

[B2]

How often have you found rigidity to be an issue?

I just saw this at the bottom of your earlier post. Sorry I did not respond. I am probably not the best person to answer this question as I have not done any big jobs yet. I seldom do. Most of my work pieces are small.

However, I will say the following. While leveling the lathe I found that when I adjusted 3 corners for the bed the 4th was commonly hanging in the air even though I was near level. Then when I would lower the hanging corner any more than just making contact the floor one of the other corners would lift. That is the bed was so stiff that it did not seem to twist enough for me to observe it even with the weight being only on 3 feet or almost all on three feet rather than four. During this I had the tail stock and tail stock stand etc weight hanging on that end of the lathe. The saddle was in the middle of the ways. When leveling the lathe I first tried to level the head stock, but there is no clear surface to use for measurement. There are 4 leveling screws/feet under this end and 2 under the tail stock end. There maybe a better way to do the head stock level, but my approach was simply to put a long bar ~1.25" dia. steel bar in the chuck and put a level on it and then to start adjusting the screws. To get the front to back level I then put the level on the bed ways at the spindle. My 90 year old concrete floor was intentionally poured to differ by almost an inch in height over the length of the lathe for draining with a little front to back slope as well. The lower floor height is at the tail stock. In the z-axis I tried to match the level readings I had gotten on the spindle chuck bar etc. So the leveling bolts at the tail stock are extended out ~ an inch longer than the head stock feet bolts.

I also wanted to mention, I find the big scale on the handle for the z-axis to be confusing at best. It starts at 0 and goes up to 0.58" for one turn. So once over that it starts at 0.0 again. This means that if you want to go more than 0.58" then you have to add the next rotational readings to the 0.58." This must be repeated for each turn. This is confusing, but not impossible. However, when you are going the other way you have to count down from 0.58 and then add that amount to the amount you have already traveled. While math comes easy to me, I do not practice or naturally do counting down from 0.580" on a hard to read scale and then add this to 0.580. So the solution to this is to get a DRO and never look at that wheel scale! Besides, mine is off by 0.004" per turn anyway and this would have to be compensated for! The z-axis the lead screw is 4TPI. The cross slide is fine, 10TPI and 0.1" per turn as mentioned in the literature.

Dave L.

 

[Scott-ak]

I have not experienced the jumping out of gear. I have had plenty of times where I did not have it properly engaged. That occurs very seldom now after some time with the machine. As for changing the threading/feed gear box the only time I've had any real trouble is the C-D when changing for large pitches, 6 tpi etc. Even then it only takes bumping the jog and it goes in/out fine. I use the VFD for phase conversion. I simply don't see the need for all that other stuff. The gear train provides all the flexibility needed. The only time I've thot finer control of rpm might help is when picking up a thread for a repair and 50 rpm feels fast. I get it, some folks like to tinker w/tech and all the bells and whistles are cool but I don't see them as helping turn out a better part. I don't think PM machines are gods gift to machinists. They certainly have their shortcomings. Use'm for what they were designed for and it should serve you well. Trying to make it something it isn't just leads to frustration. Einstein said something like 'everything should be made as simple as possible and not one bit more'.

 

[B2]

Thanks Scott-ak,

I appreciate your comments, observations and perspective. I will figure out the gear jumping some time, but I do find it hard to get the speed gear into the M position and keep it there. I just do not feel the detent. A little to one side or the other of the M position and the spindle just free wheels (does not engage at all). This would probably be ok, but is also dependent upon whether one is shifting from H to M or from L to M. I suspect I will eventually get use to this and it may start to come naturally. Or, I may I just need to tighten the detent ball spring more!

Yes, phrases like "keep it simple stupid" and "measure twice and cut once" do not go wasted on me. And yes, I agree, bells and whistles do not make better parts or necessarily enble making them faster. However, I have no real regrets in adding the extra features to my VFD conversion. They sort of come with the VFD and I had to do all the work anyway. So far I have found all of them, except the coolant add-ons, to be of value. (The coolant pump system is just messy and I seldom need much coolant/lubricant for my jobs.) For those of us, who don't do enough machining to develop proficiency can probably appreciate the proximity sensor approach to threading, or approaching a desired shoulder, more than those who cut threads regularly.

I am not really a machinist and make no pretense to be! You sound like you have the skills of a lot of practice. By trade I am a research scientist and teacher and sometimes a technical consultant. (Electronics, optics, magnetics, materials...) As such implementing the VFD conversion and exploiting the extra features of the Hitachi VFD is just sort of what I enjoy doing .... solving another puzzle, but certainly not a necessity! As mentioned earlier, I have an old South Bend Heavy 10 (love the extended threading gear box) and I have messed up more than one re-thread due to my lack of skill or the wear on that old machine. Nevertheless, I have found the SB to be very helpful at making my small parts. Nevertheless, I appreciate the simplicity of just turning on the old SB and its running the same way it did last week or month. I seldom even move the belt over to another of the three fixed spindle speeds.

I do a lot of what some folks would call strange, or unusual, experiments and so commonly need to build parts. Hence, the machine shop .... I very seldom make a second copy of any one thing, but do sometimes do the work over to achieve an improvement. Some of these experiments require unconventional parts and I use my lathe and mill to make some of these. I like having the RPM meter as I do run the motor at lower than conventional speed and it is nice just to see what the speed is. I have recently, and am in the process of, added an electronic counter so that I can quickly determine the number of spindle turns (actually number of tenth spindle turns). I have wound more than one magnetic core and being able to run the lathe at very low RPM, or to be able to jog it at very low rpm is nice. Neatly wound tight cores, even on rectangular shaped mandrills, for magnetic devices, is a necessity. On the SB I always had to turn the spindle by hand, which takes a long time and is hard work, when making 500 turns. Also, I commonly would lose count and knowing the winding's number of turns is important. When you lose count there is no easy solution... take it a part, discard the wire, and start over. Now, I will have an automatic count!

Dave L.