© all rights reserved G W Howe 2017 - 2020


Article review of the BH600G Lathe & Modifications

GWH Engineering

creative engineering in a home workshop

The BH600G is essentially the same design supplied by many different suppliers in various countries.  That said i am sure the place of manufacture varies as does the quality and although the design is similar some details vary.   These lathes offer a very reasonable lathe especially for the small home workshop and as sold they include nearly all the essential bits that are often only available at extra cost on non-Chinese lathes.  When I bought my Myford ML7 lathe in 1962 as new it came as a very basic package, no motor, no stand, no 3 or 4 jaw chucks, no tailstock chuck, no fixed or sliding steady rests, no screw cutting gearbox, no clutch and no electrical safety provision or motor switching.  Build Quality:  The BH600G is manufactured to satisfy a specific market and that largely affects the quality of manufacture.  The Myford was built for a largely home market at a time when where there was virtually no competition and was mainly aimed at the hobby machinist and so the design had many additional add-ons but all at extra cost!  Without question the Myford is manufactured to a much higher quality standard and in many respects hand built by experienced fitters.  However, the relative cost of a Myford is, in todays market at least 4 times more expensive than Chinese lathes of a similar or greater capability.  Design:  The Myford unfortunately is from a design originating in  the early 1950’s and very little was changed as the years went on.  The Super 7 provided more features like a gearbox, clutch and better bearings but overall it retained the same old design from the ML7.  By contrast the BH600G was also from a basic design but one which had taken advantage of modern production engineering techniques and materials.  For me the crucial change of design was the spindle through hole diameter as this radically improved the scope of work that could be done.  Conclusion:  I can sum this up by saying since I bought the BH600G 5 years ago I have used the Myford probably for less than 1 hour in that time.  But , in the case of the BH600G, you do have to check it out and many areas will  some require re-work or better fitting.  The major consideration when buying the BH600G for me was that it was a very simple design and in most cases I could improve the fitting of existing parts and should parts need replacing the cost was low.  The critical areas of design and manufacture of any lathe is the main bed ways, the spindle and bearings.  Fortunately, I found these are all of a good standard on the BH600G,  The ways are accurately ground and hardened , the spindle is well made and also ground so that commercially available bearings are more than adequate for home workshop use.  It is interesting to note that in modern times the manufacture of bearings has radically improved such that even standard grade bearings are of a very high standard when compared to 20 years ago.  I was told by an engineer who visits China that the machines used to grind lathe beds are very high quality and totally automated. The rest of this article will describe the modifications and remedial work I have done to correct some of the poor fitting as supplied.  Stand & Levelling: The stand unit as supplied is quite satisfactory but I fitted additional cross bars (12mm x 50mm)  to give more rigidity to the floor mounting points with extra width so that although free standing on a concrete base it has a greater foot print area.  At each end of the bars is a 12mm tapped hole for a bolt and locking nut arrangement so that levelling is easily done.  The bolt sits on an aluminium pad to resolve vibration issues.  Cross Slide and Compound Slides: Unfortunately, these are not fitted that well.  The taper Gib is a great idea but difficult to re-fit unless you start with new longer Gibs and then scrape them to a precision fit.  I did scrape them to ensure the sliding surface was flat and would have good contact.  The compound slide was a reasonable fit to start with so the scraping certainly improved matters.  The cross slide had a taper issue so I scraped the gib so that it was flat and decided to correct the taper problem by having 4 M4 adjustment screws fitted to the side of the cross slide to locate back side the gib. This is exactly the same as used in a Myford which has parallel gibs.  This is a simple and effective solution and the amount of screw adjustment  needed is minimal, probably 0.0001”.    Recently added a new quick withdraw action for the compound slide when cutting screw threads. (see link) Tailstock: The fitting of the tailstock to the ways was in my case poor.  The amount of contact area with the ways was probably 10%. The solution required quite a lot of scraping  to increase the contact are to about 70%.  That accomplished the tailstock had now lost height and so 0.010” shims were needed between the sliding bottom block and the tailstock top.  This is not an easy task and by judicious scraping eventually the tailstock is within the stated precision requirements, actually better than as originally supplied.  This shows how poor Chinese fitting can be and how they get away with hidden problems, totally unacceptable but now resolved.  Tony Mount had described in Model Engineer mag. a modification to increase the clamping of the tailstock barrel so I added that for good measure.  A much better clamp. Chucks: I was very concerned that the 3 and 4 jaw chucks supplied would be low grade and probably need to be changed.  On the Myford I bought Pratt Burnerd chucks which were made to a high standard and also quite expensive!  After much use  I have to say the supplied Chinese made chucks are excellent though the back plate fitting was poor so this had to be addressed.  Warco supplied me a new backplate for the 3 jaw as the original was a sloppy fit.  I like to ensure that when the backplate locates to the spindle register back it does so squarely.  On checking, the locating surface was not good and need scraping to give 90% contact area all round, this makes a big difference to the security and accuracy of the chuck.  The screw threads have clearance  and the register diameter also has clearance such that if the locating register back surface is out then the chuck will tilt over and cause vibrations  especially evident with large heavy chucks.  Tool Posts: The supplied tool post was changed to a new type with drop-on tool holders which can be easily and quickly adjusted for height.  This was Chinese made and all surfaces ground square with a blackened finish.  This probably one of the best buys for any lathe as tool changing is so convenient. In addition to the main tool post I designed a rear tool post based on one once used on a Hardinge lathe.  This provides a very substantial tool post which |I mainly use for parting off cutters which are inverted.  More details on this can be found here Fine Feed: For me a problem existed on the BH600G with the carriage fine feed which was noisy and not that fine.  I designed a simple belt drive from the spindle to the gearbox which is almost silent in operation and provides a range of fine feeds from 0.0017” upwards.  More details are available on this link. Carriage Stop: I read about a problem which caused the carriage to run into the chuck of a lathe causing extensive damage.  I designed my own version of a sliding carriage stop which uses a LED to illuminate when the carriage touches the stop.  This is very convenient when cutting threads but also the design incorporates an auto emergency stop to the lathe motor if the LED when illuminated but the carriage continues forward on overrun.  This is especially useful when working close to the chuck jaws and for what ever reason the carriage feed clutch is not disengaged at the right point.  The stop is positioned so that the overrun is limited to about 1mm.  More information on this is available here Screw Cutting Clutch: This is a major change to the BH600G to replace the manual screw cutting method to one as once used on Hardinge lathes.  The design for this was done by Graham Meek who had made several designs for Myfords and other lathes but not the BH600G.  The screw cutting clutch enables the task of screw cutting to be done at higher speeds and removing the need to stop the motor and reverse to start a new cut.  This is automated by adjustable stops and makes screw cutting a very enjoyable task. Details about this can be found here Soluble Oil feed: I have never been a fan of full flow soluble oil on a home workshop machine due to the mess made however it is necessary to have some limited suds feed as and when required.  i designed this pump which is foot operated to provide a small flow of oil. Details are available here. Drive Unit: The BH600G is a basic belt driven lathe with back gear arrangement exactly the same as in the Myford.  The only problem is the changing of the motor belt from large to small pulley was difficult as I had no access to the rear of the lathe.  The best solution is to incorporate an inverter drive unit which then gives easy speed changes.   Details are available here Saddle: The saddle as delivered was painted yellow! which was changed by me to black however the saddle near the chuck and to the front often gets hot chips on it and these bed into the paint.  The remedy and well worth the effort is to scrape off all paint back to the milled surface and I scraped this and finished it with emery paper to give a ground look finish.  This is so much better and easier to clean. Belt ‘clutch’: Normally, the belt is tensioned by a lever and this also enables it to ‘relax’ so that the belt can be moved to different pulley ‘V’ groove.  The belt to the motor is difficult to change and so with the addition of an inverter drive I have the motor drive set permanently to highest speed.  The drive to the spindle pulley uses a substantial belt and even in the ‘relaxed’ lever state the belt still has a drag effect on the spindle making it difficult to rotate the chuck by hand.  This is sorted by the introduction of a simple mechanism to act as a sort of  ‘clutch’  which causes the belt to clear the counter shaft pulley  thus removing any drag.  The mechanism is very simple and consists of a wood block fixed inside the headstock housing at its base, this keeps the low side of the belt at a constant height, and another strip of wood (fixed to a cross bar at the top of the housing).  When the belt lever is ‘relaxed’ the belt is held clear from the pulley so eliminating belt drag.  Wood is used to prevent wear on the belt and in running mode the belt is clear of the wood surfaces. Leadscrew End Play: The leadscrew on the BH600G has a support bearing at the tailstock end but there is no ability to take up any end play.  At the gearbox end the leadscrew is fixed to the output drive from the gearbox using a roll pin but end play is dependent on the fit of the gearbox output shaft.  On mine the end play was minimal but still for screw cutting very fine threads that can cause thread pitch to be variable.  The correction is to fit a pair of needle roller thrust bearings at the tailstock end and play is taken up by applying a very small amount of lead with an end nut.  The leadscrew shaft requires a short extension which is fixed to the leadscrew end with an M8 thread and thread loctite to ensure it is secure.  Assembly is simple and this fix completely eliminates any play.