Jump to content

timgunn

Members
  • Content count

    134
  • Joined

  • Last visited

  • Days Won

    1

timgunn last won the day on May 19 2016

timgunn had the most liked content!

Community Reputation

11 Good

About timgunn

  • Birthday 03/15/1962

Profile Information

  • Gender
    Male
  • Location
    Lancashire, UK
  • Interests
    Tools, science, food, wine. Making things that "just work".
  1. timgunn

    belt grinder motors and vfd uk advice required

    A VFD takes mains power in, rectifies it to DC internally, then synthesizes something that looks to a 3-phase motor sufficiently like a 3-phase sinusoidal waveform for the motor to behave as if it is powered by one. The clever bit is that the VFD can vary the apparent Voltage and Frequency and make the motor run at variable speed. VFDs are available for single-phase 230V input or for 400V 3-phase input. If you get a 230V single-phase one, it can run on UK domestic mains. The biggest you'll be able to run from a 13A socket will be a 3 HP/2.2 kW. It's not really worth getting any other size IMHO. They can run motors smaller than their maximum rating, but not bigger. If you use a 230V VFD, the output will be 3-phase 230V. The vast majority of 3-phase motors up to 3 HP/2.2 kW are wound for 400V connected in Star (Wye) or 230V connected in Delta and can run fine on a single-phase-input VFD. You'll need to check when ordering though. Over 3 HP/2.2 kW, motors tend to be wound for about 700V in star and 400V in Delta to enable star-delta starting (an old-school way of reducing motor starting current. It has largely been supplanted by VFDs). These cannot run on 230V 3-phase. Here in Europe (for the present), we have 50 Hz mains: 50 cycles/sec, 3000 cycles/min. Other parts of the world have 60 Hz mains: 60 cycles/sec, 3600 cycles/min. It's the reason you'll see the different motor speeds quoted on opposite sides of the pond. A 2-pole motor will run at an RPM equal to or just below the frequency of its power supply (3000 or 3600 RPM). A 4-pole motor will run at, or just below, an RPM equal to half the power supply frequency (1500 or 1800 RPM). 6-pole: one third (1000 or 1200 RPM), 8-pole; one quarter (750 or 900 RPM), and so on. We tend to use either 2-pole or 4-pole motors and in Europe, motors generally conform to IEC standards. Across the pond, they tend to use motors to NEMA standards. NEMA motors are pricy over here and offer no inherent benefit. They tend to be used where expensive machine modifications would be necessary to change to an IEC motor. For a serious belt grinder, you'll probably want a 90-frame motor in a long casing (90L). This will most likely be 1.5 kW/2 HP if it's a 4-pole or 2.2 kW/3 HP if it's a 2-pole. The shaft size of 90-frame motors is 24mm. Because half the world uses 60 Hz mains and the other half uses 50 Hz, meaning that maximum mains speed is 3600 RPM, motors are designed to run to 3600 RPM. It is not cost-effective for motor manufacturers to design a completely different motor for each speed, so the only difference between the 2-pole, 4-pole and 6-pole motors in a particular frame size is the winding. The winding is attached to the inside of the outer casing and is static. This means that all the moving parts are good to 3600 RPM. We can run a 4-pole motor to 120 Hz to get 3600 RPM using a VFD, or we can run a 2-pole to 60 Hz to get the same 3600 RPM. At the bottom end of the speed range, most drive/motor combinations are able to run smoothly down to about 10 Hz. Below this, running from the simpler V/Hz drives, things tend to feel "coggy". The V/Hz drives use a fixed (usually linear) relationship between Voltage and Frequency to determine what will be supplied to the motor and this linear relationship tends to break down once it gets that far from the design frequency. There are also drives which have "Sensorless Vector" capability. These measure the time difference between peak current and peak Voltage internally, calculate to determine the phase angle between them, then fine-tune the Voltage in real time to maintain the design angle (the motor Power Factor defines this angle, being its Cosine). These can keep the motor running smoothly well below 10 Hz and usually down to 1 or 2 Hz. A 2-pole motor on a V/Hz drive has about a 6:1 speed range (600-3600 RPM, 10-60 Hz). A 4-pole motor on a V/Hz drive has about a 12:1 speed range (300-3600 RPM, 10-120 Hz). Either motor has a much greater speed range when run from a SV drive with smooth running down to 1 Hz achievable if needed. VFDs switch large currents very fast and produce some heat, which must be dissipated. Most VFDs have ventilation fans and allow airflow over the power components to cool them. They are intended for use in clean conditions (usually sealed electrical enclosures). If there is airborne steel dust (which is both conductive and magnetic) it will flow across the power electronics, where the magnetic fields caused by the switching will capture it and attract the metallic dust right onto the power components. The short-circuit that results is usually quite spectacular and is invariably expensive, killing the VFD completely. If they are going to be in the same room as a grinder, we need to use VFDs that are protected against such dust to IP66 or NEMA4 standards. We can either mount an unsealed drive in a sealed enclosure ourselves, with sealed control switches and speed control knob, or we can buy a drive that is designed to be sealed to IP66 from the factory and which has the sealed control knobs, etc on the front. The latter is by far the better option for the non-electrician. To buy and enclose an unsealed (IP20) drive properly, with sealed controls, to IP66 is about as expensive as buying an IP66 drive to begin with. In the US, the KBAC series of sealed drives from KB Electronics are the go-to. In Europe, the Invertec IP66 drives tend to be the ones people use for grinders. The current ones are SV drives so you get the low speed capability. I'd recommend a 90L-frame motor and an Invertek ODE-3-220105-1F4Y VFD. The drive is expensive, but it's a cry once thing. I'd try for a 2 HP, 4-pole motor for personal preference, but would be pretty happy with a 3HP, 2-pole. If you are anywhere near Lancashire, I can sort you a foot-mount 90L 2-pole 3 HP from a compressor, gratis.
  2. Hi Tim,

    Knowing of your expertise, especially on AMAL gas injectors, I wonder if you could please advise on my comments below? 

    I have built a forge from Vitcas Grade 28 Insulating Fire Bricks; the aperture (as originally constructed) measures 7.5” w x 4.75”h x 18” l (195 x 120 x 460) and have been using it for a couple of years without problems.

    Forge Performance Test.jpg

    I use a ¾” AMAL injector (on a burner made to the well-proven design) by simply locating it into the front of the forge to heat the forge where required. For forging mono steel and pre-made san mai this works fine and I have successfully forged blades in a wide range of carbon and stainless steels.

    I recently hand-forged some san mai (140mm x 35mm x 3mm pieces) from spring steel/mild steel as a test piece and found that there was incomplete fusion in some small areas; I put this down to insufficient heating but it could of course be poor hammering technique. I don’t have a power hammer as I can’t stand the noise and our neighbours certainly couldn’t! But, the attraction of forging my own san mai and Damascus is growing so I’ve given thought to my forge’s capability hence the performance test. Additionally, I am contemplating building a hydraulic press.

    In addition to the AMAL burner, a year ago I bought a ¾” T-Rex burner from Hybrid Burners to evaluate against it, I concluded that the T-Rex has no performance advantage over the AMAL.

    So, last weekend I modified my AMAL by adding the support tags, bored a hole into the forge top, reduced the forge volume (only by blocking which is not substantive but guess the reduced length helped) to 6.25” wide x 4.75” high x 12” deep (155 x 120 x 305) and then ran some tests the results of which are graphed  below. I monitored the temperatures with a thermocouple the tip of which just protruded into the top of the chamber. 

    image.png

    As you see, the maximum temperature I eventually reached was 1220 at a gas pressure of 1.75barg. The latent heat in the blocks kept the temperature quite high even after I had reduced the gas pressure; I ended the test at 80 minutes.

     So, based on your knowledge of these things: 

    • Is this the performance and temperature I should expect from the AMAL with its standard 0.036" jet in this size forge
    • Should 1220 degrees be sufficient to allow forge welding by hand or do I need to higher temperatures and if so how to get there - fit 2 AMAL’s ....
    • What’s your view of modifications to the forge for an optimised forge design  

    This is such a wide area that I hope these few questions will set my development path and would appreciate any comments or suggestions.

     Clive Witton 

    www.instagram.invictaknives.com/ 

  3. timgunn

    Spring rate

    Given that a failing hammer can also take parts of the operator off, I'd feel happier with: yes, mild would have worked in that application just fine as long as the fatigue limit didn't get exceeded.
  4. timgunn

    Forklift tine anvil

    What with? With 2 cuts and probably some tidying up with a grinder, it should be less than a days work to make a better post anvil than many smiths are ever likely to own. That seems like an excellent use of your time to me.
  5. timgunn

    thermocouple recommendations?

    I've had pretty good results with a few of the cheap Chinese pyrometers. I have access to a thermocouple calibrator and always check the accuracy of the pyrometers across the intended working range before using them in anger. Until a couple of years ago, I used to recommend the TM902C, available for around 5 bucks delivered. I'd had maybe 20, ordered in ones and twos over maybe 4 or 5 years. They were boringly accurate from 0 degC to 1370 degC (32 degF to 2500 degF) and came with a glassfiber-insulated bead probe that was good to 400 degC (750 degF) and was flexible enough to be closed in an domestic oven door to check tempering temperatures. A couple of years ago, I ordered ten of them. When they arrived, I checked them on the calibrator and found they were accurate up to 800 degC but then became progressively inaccurate. I can't remember whether they showed 1370 degC at an input of 1290 degC, or showed 1290 degC at an input of 1370 degC: either way I was not happy with them. They were very consistent and all read the same on any given input signal. Side-by-side, there were external differences between the good ones and the bad ones, but these would not be obvious on an ebay listing so I stopped using or recommending them. I have since used DM6801 digital thermometers and found them boringly accurate. 10-20 bucks delivered from China. The supplied bead-type probe is PVC insulated and only good to a little over the boiling point of water: not much use to most of us. Whatever pyrometer you get, the supplied thermocouple will not be any use in a forge and you'll need to go to a thermocouple specialist. The "best" probe I have used for "our" purposes is an Omega KHXL-14U-RSC-24 which is a handled type K Mineral Insulated probe 1/4" diameter and 24" long below the handle. The sheath material is Omegas proprietary "Super Omegaclad XL", which is not really necessary for Heat-Treating, but seems to be the best I've found for surviving being used to occasionally check forge temperatures without going to painfully expensive Type S, Platinum-based, thermocouples. The assembly includes a curly cable and miniature plug. https://www.omega.com/pptst/KHXL_NHXL.html Using the part number builder on the page does not allow anything longer than 18", but putting the desired part number into the search box, top-right, will get you a price. For sword-length stuff, a 36", 48" or even 60" might even be worthwhile. I am cheap and tend to use similar thermocouples from the company we use at work. However, these "only" have a type 310 Stainless steel sheath, which is rated to 1100 deg (2012 degF) and will last pretty much indefinitely for HT, but doesn't last many cycles to 1300 degC-plus when used for checking welding temperatures are being reached.
  6. timgunn

    Kiln - to buy or not to buy

    A Kiln/HT oven will certainly allow you to do things that no forge can realistically manage. Power input is typically around 3 kW. I have built HT ovens up to 27" long that will reach 1300 degC (in about an hour and 45 minutes) and will run from a standard UK domestic outlet: 13A at 230V nominal rating, so 3 kW. Bigger ones might use more power, smaller ones might use less. Lower temperatures are reached much faster. When I first tested the 27" oven from cold (maybe 10 degC, 50 degF), 800 degC, 1472 degF was reached in 22 ½ minutes, 1100 degC, 2012 degF took 54 ½ minutes, the temperature at an hour was 1125 degC, 2057 degF and 1177 degC (2150 degF) took 71 minutes. As Brian says, once the desired temperature is reached, the PID controller will cycle the power on and off as needed to maintain temperature, so the average power drawn during the "hold" will be less than the heat-up, which will usually be at full power unless you are "ramping". A fairly typical HT cycle for O1 might be to heat the oven to 800 degC and stabilize it (call it 30 minutes at 3 kW, for 1.5 kWhr), triple normalize; Work in, allow the temperature to recover, soak for 10 min, pull out and cool to black. Repeat. Repeat. (Call it an hour at 33% power, so 1 hr at 1 kW for 1 kWhr) Then Austenitize: work in, allow temperature to recover, soak for 20 min (call it 30 min at 33% power, 30 min at 1 kW for 0.5 kWhr). Quench workpiece(s) and switch off oven. Total power consumed 3 kWhr. You might expect to maybe double that for a Stainless HT, but to put things into perspective, unless you only do single blades each time, 6 kWhr is still unlikely to significantly exceed the cost of HT foil. The numbers I've given are probably high-side figures. I don't have hard data, but I'm pretty sure the same homebuilt 27" oven was actually showing a steady-state output cycle of only 18% during the 1100 degC, 2012 degF hold section of a Stainless Heat Treat: when I delivered it, the new owner gave it a test run while we drank tea. I use a 2-second output cycle and always have an LED that is switched with the elements. I was surprised the "on" part of the cycle seemed so short and pulled up the output cycle display on the controller. It read 18%. The heating load of a HT oven is entirely resistive: it does not have the starting surge of a motor load or the horrible Power Factor of some modern electronics. So long as you do not exceed the outlet rating, I would not expect any new and interesting problems with mains power supplies, even in the middle of nowhere. The short-cycling 3 kW output would be harsh on a small generator, were you to run it from one, but I'd not anticipate a problem once the generator gets above maybe 10 kW or so.
  7. timgunn

    Grinder and VFD Programming

    The first question is probably "what does the motor rating plate tell you about the motor?" and the second is "what do you want it to do and how do you want to tell it to do it?" The motor rating plate should give you a bunch of values that need to go into group 2 of the programming. The values you are likely to have are 02-01, 02-03, 02-04, 02-05, 02-06. If the motor plate gives you any others in an obvious way, put them in. Otherwise leave the settings as the defaults. The default command source seems to be the keypad. If you have a speed control potentiometer (twiddly knob) on the front, set parameter 00-05 to "1" to enable speed control from it. It looks like it should run on the start and stop buttons on the keypad and should go to whatever speed the knob is set to. Once running, adjust the knob to adjust the speed. It looks like the default acceleration and deceleration times are 10 seconds, so there will be a lag if you turn the knob fast. The drive is environmentally protected to IP20: effectively fingerproof. It is not sealed against dust and metallic grinding dust WILL kill it very quickly unless you take steps to seal the drive. Bear this in mind when you first fire it up: do not get over-enthusiastic, start grinding and cause fireworks. The quickest anyone I know personally has killed a drive this way is around 2 minutes from first switching it on. You'll then need to decide how you are going to protect the drive and how you are going to control it in whatever protection you come up with (if it's in a box, you won't be able to get at its front panel). You can also decide on the speed range you want. Some of the remaining parameters will set the drive up for the speed range and command source you decide to use. Many of the others are there to interface with control processes in industrial automation and will not be needed for a grinder with a human operator. I mount IP20 drives in IP66 enclosures and fit a remote control box on a trailing lead with start and stop buttons, a keyed forward/reverse switch (the key is taken out when running a machine that should not be reversed) and a speed control potentiometer. The box has a 3-phase socket on the front so I can plug in different machines. It's usually as cheap to buy an IP66 drive to begin with, and is much less effort.
  8. timgunn

    I think I bought the wrong burner

    It looks like the burner is choked down to me: The air holes are covered so there is not much air flow for the gas to burn with. Try changing absolutely nothing except the choke sleeve position: move it towards the forge and uncover more holes to get more air in. See what happens and let us know. If you can, take photos in the dark. The most useful shot is often one across the mouth of the forge to show how much Dragons Breath there is, and what colour it is: In daylight, the DB flame does not show particularly well and digital cameras tend to do strange things to the white balance so that the flame colour in the photo is not the actual colour. Taking a tight shot directly into a forge can be completely useless because the camera will adjust the colour and a real dull red will look in a photo just like a real bright orange in another photo.
  9. timgunn

    Heat Treat Foil

    Are 17-4 and 18-8 foils actually that much more readily available at .001"-.002" thicknesses and in knifemaker-type quantities? 321 is not all that far off being an 18-8 stainless: it's effectively 18-10 with up to 0.8% added Titanium, which apparently improves high-temperature stability. I think 321 was developed to counter cracking in aircraft engine exhausts. Neither 304 nor 316 are particularly recommended for high temperatures. 309 is higher in both Chromium and Nickel for use at higher temperatures. They get used because they are known to work reliably in their specific applications. "Someone" did some work to identify suitable materials for the tasks and that is what they came up with. Most folk seem to accept the consensus view and pony up for the proven product. Using 321 when heat-treating steels that "need" 309 has been known to fail and I'm pretty sure an evening spent searching the knifemaking forums will provide some insight into the sort of things that can go wrong: I'm pretty certain I've seen posts about the foil welding to the workpiece and to failure of the envelope allowing contact of the workpiece with air. If you have easy access to other grades of SS foil and can do some testing, I'd certainly love to see your results.
  10. The Auber stuff should be OK. I'd forget the element you linked to entirely. It is waaay too thin to last. I built my first 5 ovens using 16AWG (1.29mm diameter) elements wound from Kanthal A1. These were ok-ish: fine for occassional hobby use, but a couple of the guys who use them to put food on the table had element failures. I then went up to 1.6mm diameter Kanthal A1 and they seem to be lasting better. I got some of the Far-Eastern elements off ebay, just because they were so cheap that I had to check them out. They really are very thin indeed. They also appear to be continuously-wound and then cut from a roll. To get connection tails, you'd need to unwind some of the coil and twist it to make the tails. I've just dug one out and measured it at about 0.65mm diameter wire. My advice is to go no thinner than 16AWG, thicker if possible, and to use Kanthal A1 or equivalent. The price will, of course, be considerably higher than that of the thin ones on ebay.
  11. Give it a try and see if there is a problem. As you will have a regulator in the system already, a local shutoff valve and a needle valve to control the flow to the forge would probably get the job done.
  12. It will "probably" be fine, but with caveats. Inside the regulator, a plug moves in and out of a hole, controlling the amount of gas getting through. The plug is attached to a diaphragm and this works against a spring. The gas, downstream of the plug and hole, presses on the diaphragm and spring. As the pressure rises, the plug moves into the hole. As the pressure falls, the plug moves out of the hole. The steady state is where the pressure on the diaphragm balances against the pressure of the spring. Adjusting the spring preload adjusts the regulator pressure. The size of the plug and hole are important. These are fixed in your regulator. If your regulator is rated for use with, for example, 50 PSI upstream pressure and 5 PSI downstream pressure and will allow 5 lb/hr of Propane to flow at those pressures, the pressure available to drive the flow is 50-5=45 PSI. The regulator can control the flow to less than 5 lb/hr by moving the plug into the hole, but once the downstream pressure drops below 5PSIG and stays there, the regulator will be fully open and the size of the hole is what matters. If you now supply the regulator with 10 PSI gas instead, the pressure available to drive the flow is 10-5=5 PSI. Only 1/9th of the original pressure. Flow through an orifice varies as the square root of the pressure, so the maximum throughput the regulator could provide would be the square root of 1/9th: 1/3rd of the design flow or 1.67 lb/hr. The numbers I used were chosen because they are easy numbers. However, the vapor pressure of Propane at 30 degF, just below the freezing point of water, is 51 PSIG. At -20 degF, it is 11 PSI. Both are similar to my easy numbers. The effects of Propane cylinders "freezing" are well known/documented in smithing circles. Be aware that you may find yourself seeing identical symptoms to cylinder freezing as a result of using a, now too-small, regulator fed from a 10PSI supply.
  13. First question: what do you want to HT? If you are absolutely sure you are going to need Ramp/Soak, go for a Ramp/Soak controller. If you are unsure at this stage, go for a basic controller. You'll have a pretty steep learning curve ahead of you (unless you are already familiar with PID controllers. This seems unlikely because you are asking the question). NEVER buy a controller until you have downloaded the manual from a non-password-protected site, read it thoroughly and understood at least most of it. The reasoning is this: you have a steep learning curve ahead of you. If you need to ask for help on this, or another, forum, there are likely to be some folk with enough experience and knowledge of Process Control, and Controllers generally, to help you. However, they are unlikely to be intimately familiar with your specific controller. Being able to provide a link to the manual gives them a reasonable chance of being able to provide specific assistance. Some manufacturers put access to their manuals behind a login/signup page. I tend to give up at that point. Manuals are difficult and expensive things to write. A manual translated from Chinese into English by an online translation program is not likely to be very helpful. Pretty much the entire value of a controller lies in it being able to control your process. If the manual is not adequate to allow you to set it up to achieve this, the controller is worthless. Industrial controllers, particularly the ones with higher-end capabilities like ramp/soak, are intended for use by Process Engineers: people who understand both their process and controllers in general. Programming ramp/soak profiles is usually a pain in the neck. The ramp/soak programming is usually set at the same access level as the input type, P,I & D parameters, etc. They are not like the kiln controllers found on the big-name kilns, which are designed to be programmed by an end-user with no real interest in the finer details of process control. These have different access levels for the commissioning technician and the end-user, limiting the amount of damage that can be caused by an incorrect button press. Ideally, you want a controller with the capability to do what you need it to do, with a good manual, written by someone who writes manuals for a living, in your own language (though if Swedish is your mother tongue, your English certainly seems good enough that you could cope happily with a manual in English). Then you want "real" technical support in a time zone you can live with, also in a language you can communicate in. For me, it needs to be telephone support (ageing Luddite and one-finger typist) but YMMV. I seem to vaguely recall a few HT oven builds using the linked controller on different forums. If the manual is ok and available online, it might be a good choice. Alternatives, if you are set on ramp/soak, are the Auber Instruments SYL 2352P, which I have not used myself, but which seems to be highly regarded Stateside. Support is reportedly very good. My personal favourite is the Automation Direct Solo SL4848VR or the Omega CN7823. These are, as far as I can tell, the same controller with different badges and I buy whichever is cheapest at the time. I have used them on all but the first HT oven I built (7 out of 8 so far). Support from both suppliers is superb. Omega are about the biggest name in temperature control worldwide and their technical sales people are extremely knowledgable (and patient, IME): well worth a phone call if you are unsure about thermocouples, etc. If you can live without ramp/soak for the forseeable future, a simpler controller really could be a good plan. If you go for a 48mm x 48mm (1/16 DIN) controller format and leave enough length on the wires to reach a different terminal layout, upgrading to a ramp/soak controller later is simple.
  14. timgunn

    Wiring a kiln

    Auber seem to be highly regarded Stateside. They do not have a UK presence, so I have not used them myself. AutomationDirect and Omega are both very good and have knowledgeable support staff at the end of the phone (I'm a Luddite, but they are probably great online too). IME they are also remarkably patient. If you are going to talk to them, make sure you have all your settings noted down and try to appreciate that, whilst they do know their equipment, they do not know the specific process you are trying to control: it is up to you to tell them exactly what the controller needs to do. If you are not sure about your process, a call for help on the forums may be the better first step. You are likely to find someone who has done what you are trying to do and can get you up to the point where you know enough to have a productive conversation with tech support. My favourite controller for homebuilt HT ovens is the AutomationDirect Solo 4848VR or the Omega CN 7823 (same controller, different badges. I buy whichever works out cheapest at the time) with a DC output to drive an SSR and ramp/soak capability. Bear in mind that ramp/soak profile setting is always MUCH less user-friendly on industrial controllers than it is on the likes of Evenheat's and Paragon's HT ovens. For temperature control, you are likely to need a thermocouple. Omega are about the biggest supplier of thermocouples worldwide and it is well worth spending a few minutes with a notebook and a telephone picking their brains.
  15. timgunn

    Forklift tine anvil

    2”x4” felt too small for a face. With hindsight, I should have left it. I welded on a couple of offcuts from the tapered end to give about a 4” x 5” face, but welding it ruined the original tine HT. It seems really soft and the plan is to hard-face it as soon as I can source some suitable rods. I made a cutlers stiddy from another short piece of the tine and it seems to be usefully hard with the original HT. I think a tine the size of yours would make a pretty good post anvil if you don’t overheat it when cutting it.
×