Is there any special precautions before welding on my dozer? Do I need to turn the battery disconnect to the not connected position or disconnect the batterys Just want to confirm and not cause a problem Thanks ahead
Always best practice to either disconnect the battery or main disconnect switch when welding and also keep your ground close to your work.
It's never wrong to disconnect both battery cables especially on newer machines with more electronics. I have a former plow truck and there are multiple stickers saying to disconnect both battery cables to prevent damage to sensitive electronics. I also worked at a shop building winch trucks. On those trucks you had to disconnect both cables as well as unplug the harness to the ECM. Better to error on the side of caution.
On a G series dozer without a doubt turn the battery switch off and just for good measure disconnect the batteries. G's are very finicky on voltage across the electrical system.
I never disconnect. Have never heard a real explanation as to how that does anything. Millions of trucks used as welding platforms daily. Been at this a while, welded on 1000's of cars trucks tractors buildings trains and a couple planes. Thousands of cars worked on in muffler shops daily and wouldn't want someone fooling with it all.
Then you’re very lucky. I have fried 2 ECM’s, one BCM on a HD bike and a regulator on my backhoe. Now I’m a bit aggravated at disconnecting battery cables but just a couple of months ago, I was welding on a 2003 Chevy 1500 plow truck and disconnected the positive and welded a cross member in and when I went to reconnect the positive, I somehow got the wrench in my hand, between the positive clamp and the head wall and got a bit of an arc. Truck turned over fine but wouldn’t start. Screwed around with it for an hour and finally had to call my brother. He hooked into it and was stumped for a bit (GM Tech) finally had to disconnect again and do a reboot. Then there was the time I was welding on a Cat 350 Reclaimer and put the ground on the frame legs and the arc went through the drum bearings, as I was welding on the drum. That was an expensive lesson.
It's just not worth taking a chance that nothing will get fried or arced. It only takes one time to have an expensive mistake.
On my JD 290G LC, you disconnect both leads on the batteries, disconnect the ECM (3 connectors) and unplug the master controller behind a panel in the cab (6 connectors). Also, never weld with current going across a joint - always put you ground clamp close to your weld on the same member. I don't know how much electronics are on a D4G, but it is prudent to err on the side of caution.
How exactly does this protect anything,,, and like the post above, shorted a wrench while at it. I must be lucky then,,, 1000s of cars. Brand new pickups used as welding trucks every day.
Some welding trucks the decks are insulated. I think it's a dis-service to suggest that no precautions should be taken when welding on equipment with electronics. Better safe than sorry.
The "we've always done it that way" mantra does not necessarily make it right IMHO. When an ECM can cost upwards of $4k it makes sense to disconnect a few cables, just in case.
How many are messed up with disconnect reconnect. How many batteries blown up that way. As usual I ask,, exactly what are we protecting and bSic answer is better to be safe,,,, so ,, we really don't know. This is about like torquing a dry bolt,, that somehow spread to be gospel and the difference between an oiled bolt and dry should be obvious to anyone has turned a wrench.
As for insulated bed,,, it might be so but I never heard of one. Welders even bolted to these beds.
My perspective would be different if it was a hobby. Outside of the Internet I have never seen this happen and my estimation of 1000 cars is conservative. Got bud has muffler shop, would spend a lot of time unhooking sensitive stuff to do this daily. I wouldn't want someone under the hood unhooking a bunch of stuff for routine job.
When Kenworth tells you to disconnect both battery cables and unplug the ECM on a $600,000K plus winch truck or it voids the warranty only a fool wouldn't comply. I'm sure most equipment is the same. I'm also sure there have been instances of serious electrical problems from people welding without following the factory or attachment guidelines. That's why it's a warranty requirement. It's not like it's a major struggle to disconnect the battery and/or unplug the ECM. If you refuse to let a shop unplug what is required for warranty you can't go back on them when you need to replace the $4000 ECM. It's dumb to even have this discussion. Hopefully members of the forum will heed what the factory says.
Never worked in mining I tale it.? Either way carry on doing whatever you do in your own shop. That is your prerogative. You do however appear to be the only voice against.
Here’s another perspective, if I can’t trust someone to unhook and isolate sensitive electronic components properly, I wouldn’t let them touch my equipment in the first place. There’s no contest to win here. Nearly every manufacturer recommends isolation before performing any welding on the machine, they can’t all be wrong. As for what you’re changing by disconnecting the battery or opening a ground disconnect switch, you’re creating an incomplete circuit which makes it far more difficult for welding current to cause damage. Probably not impossible but definitely safer.
I haven't ever found anyone can explain why , obviously can't here either. I don't doubt kw says this. Didn't say they didn't but have NEVER seen a real explanation other than someone said so.
Seems simple to me but I guess we all require a different level of understanding to be satisfied with the answer.
Not everything in the world is a 600k KW either. This sounds like a car forum where a guy answers every oil change question. , says so in my BMW owners manual when someone asks about a lawnmower. I was being conservative,, it's prolly closer to 3000 cars and trucks. I also have been over this subject quite extensively and consult couple engineers.
Internet has made it so easy to simply repeat something that they all must be the same and if a guy can write a little he must know what he is talking about about. Open any manual, take a motorcycle. Drive first 500 miles at 50 mph. Must not have been important when it came off the factory floor. Dyno thru the gears and lay rubber.
I think one factory manual said it, if that one did we should and now the next and they simply copy. Now we got the net and we can repeat it some more till it's fact and STILL can't explain it.
I am willing to learn something about it. Tell me more?
If you have ever welded, you can see if you have a sketchy ground, when you start your arc, what happens? The ground sparks a little and makes a complete ground. Try welding with the ground going through a bearing. If you don’t think that the ground arcs through the balls of the bearing to the race, to the shaft, there is no discussion to be had with you. Do you think I’m lying about having to replace an ECM or a BCM after welding on a bike or my own personal truck? You do you and I will do what I have learned from. Lets not argue.
Right in this thread, guy manages to short circuit s battery, wonder how much damage he would have done had he left it alone?
No one said not to follow basic precautions and ground to the piece you are working on.
No but if you manage to ruin several things then maybe shouldn't be working on it. I have managed to do it 50 years without burning it up,, who Would a guy want working on it?
I been at this a while and specialize in it to some extent. I read almost everything can be found about it at one time. I prolly still weld a car or couple a week, would make that a hundred times a year, a 1000 in 10 years. Had a streak of maybe 10 a week for quite some time. My bud owned auto shop and another muffler and hitches mostly on current models and I believe GM might say unhook in their service also , kind of sounds like a generic blanket inclusion. So,, my luck is the best and yours is the worst.
2 nd sort of. This leaves as assumption that we know what happened and doesn't really prove there would not have been different result.
Keep going, you’re getting pretty good at being an individual that argues with himself quite well. That and tells big whoppers.
I went to welding school at the college when I was in HS. The guy was kind of a crappy welder but really knew men and work and he used to include a few things not in course work aimed at guys he understood and still remember him in short lecture talking about welding thru bearings,,, made sense then and I follow that since and train my help about hooking work lead to same part being welded and I am super aware of ground loops in any welding. It's instinct for me , maybe habit but I do it without thinking about it.
Now did I say you were full of ****? You start with telling me about bearings and arc like we are amateurs. You the one burn up both motorcycle and truck.
Took an arc welding class at the local community college a few years ago. I needed to get proper instruction for fixing various farm, truck and trash equipment. Many students were looking to become pipe welders which according to the instructor, Terry, is some of the most serious welding to be done. with X-rays being used to inspect every weld for weaknesses. I remember him emphasizing over again how to keep your ground close when working on trucks, tractors or anything with a gearbox. He said that differentials and bearings will arc and seize. Keep the ground at your work piece he would say. Also remember the battery disconnect thing but never was told to disconnect ecm. Over the years since I have welded on a couple items without disconnecting. the battery and haven’t fried anything yet. Take this as the most unscientific anecdotal account in the thread.
A real concern I have and do consider is making sure the body ground isnt in the welding circuit. I bet that has been an issue over the years. I have caught myself in that a couple times but never burned one off.
Been down this road setting up a couple fab shops with interconnected metal and benches too,,, bring that up on a forum and see how that goes,,, ha Guys that probably know better jump in on that without really considering some basic stuff.
You must not be doing much other than typing multiple posts on the same subject trying to convince people they don't need to follow what the OEM's say. It only takes one time for an errant current from welding to fry something. I'm sure the OEM's have had experience with things getting fried from welding and that's the reason for the warnings. Disconnecting the battery isn't a big deal.
Go ahead and tell me something about why it helps to disconnect the battery. Seems like it wouldn't be made up and should be logical. The comment that I must not have worked in a minor is interesting. But doesn't answer this or anything else. I realize not every welder has examined this or read some mine, OSHA and I and electric code in this regards but all really based and described in a couple of them Welder is sposed to know and understand,, the main point in all of them is to avoid using system wiring as conductor for welding currents and some of it rather specific and takes a while to get what they mean.
If this was an issue we would see forums with dozens of posts. Most pretty short lived with common catchall phrases.
Here is why this thread interests me and most of the net for that matter. I don't think the instant speculation matches the reality. Of all the dumb crap I have done to equipment this is pretty low on the list
If all the people said do then why should be a fairly easy answer
When i was an apprentice, the foreman made us disconnect the battery and modules. This caused more issues with bent pins then it saved. From then on I disconnected batterys. Another place I worked at used a surge protector jumper lead that went between the positive and negative. I argued with the old timer about disconnecting the battery (basically identical to the conversation above). I started to use it and never fried anything so I stopped disconnecting batteries. Now days (as a cow chaser and part time spanner handler) i just clean the earth point and put it as close to the welding area as possible. And dont strike an arc on anything thats not directly connected to the item earthed or to be welded. Its not hard. The current has to flow somewhere. If you earth the the chassis and strike on the ecm, what ever in between is going to pop.
This thread has run its course. If people with newer equipment with electronics want to take a chance and not disconnect the batteries when welding on it go right ahead. Don't come back later asking why you've got a bunch of electrical gremlins. The fact is some electronics have been damaged by welding. That's why OEM's give the warning.
Electrons behave like hydraulic oil. Each flows through a circuit because of pressure and the circuit is designed for a max operating pressure for longevity. It's generally not the operating pressure that kills circuits (whether hydraulic or electric), it's from pressure spikes caused when rapidly starting or stopping the flow that causes issues. These spikes can easily surpass the max operating pressure of a system which is why hydraulic cylinders are designed with special valves and buffers at the end of travel to minimize spikes from sudden stopping/starting. You know the spark you get when you pull the toaster plug out of the wall when it's toasting? That's back-EMF and it's a reverse pressure (voltage) spike so high (way above operating voltage) that it actually creates an arc in open air. Now think of a welder with all that same starting or stopping of current. There's lots of EMF and voltage spikes going on that could easily surpass an electrical circuit's max rating. And from my very basic understanding of welding, a MIG welder starts/stops the welding many times a seconds as the wire touches the metal to complete the circuit only to melt back a bit opening the circuit before the wire touches the metal again to complete the circuit. That rapid on-off-on buzzing sound is why MIG welding sounds so different than stick welding which is more of a hiss. And I'm not not even considering the HF used to start welds on modern welders or even used for aluminum jobs which is even more devastating to electrical circuits. Anyone that wants to weld aluminum with critical circuits attached, or even in proximity to circuits like computers without proper shielding, is just asking for damage. Sure, most electrical circuits and components are fairly isolated and probably won't be affected by welding since they are not in the direct circuit...by why take that chance if you can help it? But I also agree that batteries shouldn't be affected by welding spikes/noise either. The only reason I can think why batteries are recommended for disconnection is to physically isolate as many electrical circuits from a chassis as best as possible. Even though starters and a few other systems use mechanical relays for switches which physically isolate them from being a complete circuit, there are too many small, control circuits that use solid-state switches that are always connected inline with the batteries and thus can possible conduct voltage spikes. These are all the small gremlins that Dave is referring too because they are so hard to troubleshoot. Disconnecting the batteries helps isolate all these little circuits. And yes, most automotive grade electrical systems for critical functions are overbuilt in a highly regulated industry for safety reasons and probably won't be affected too much by welding but I wouldn't be so positive the same engineering is put into the not-as-regulated systems like heavy equipment...especially the cheap import stuff. Even the average home has issues with water hammer if the plumbing circuit is not engineered or installed correctly. It's the exact same issue from pressure spikes when water flow is suddenly started/stopped...not the operating pressure. To each their own on how they want to weld on their equipment, especially with all the modern electrical doo-dads.
Very good explanation. Yes short circuit mode in Mig has the current going on and off at a very high frequency but it's not considered a high frequency circuit. Hopefully not too many are doing heavy repairs with short circuit Mig. It's not the strongest and isn't approved for structural welding. Done fairly hot might be OK but stick is much better. High frequency used mostly for Tig on aluminum is known to cause lots of problems. I worked at one place where the high frequency took out the volt and amp meters on the welder just because there were still a couple wraps of cable on the hangers on the machine. It's also known to screw up radio's/electronic equipment and people with pace makers should stay clear. Speaking of high frequency, (the new) inverter welders operate on high frequency. That's how they are able to be so small compared to transformers. Not worth taking a chance to find out how bad they can mess up the electronics on newer machines. If someone is using an AC stick welder there could be potential problems too introducing high amp AC current to a DC system. If someone still doesn't think it's wise to disconnect batteries it's their own fault if something gets fried.
All great info, Dave. If I was to say your words in a slightly different way, I would say it was probably fairly safe back in the old days to stick weld on equipment as long a person firmly attached the ground directly on the piece they were welding to prevent arcing across bearings and other joints. The old transformer stick welders didn't really produce a lot of noise that could damage hard-switch contacts and thick wired glow plugs. Other than a couple diodes in alternators or rectifiers, I don't know of any solid-state devices in most machines from the 70's or older. Maybe even through the 80's. But fast forward to today with machines having oodles of computer chip ECMs, solid-state sensors, micro-electrical-mechanical systems (MEMS), liquid crystal displays, etc. and multiple that times the compact, high-frequency, multi-mode welders commonly used and there's bound to be problems if precautions are not exercised. Btw- I didn't know MIG had that short circuit mode and it wasn't allowed for structural welding (you can tell I'm not a welder). But I do clearly remember way back in the day when my dad purchased an old used Miller TIG welder for our shop. When he was teaching me to weld with it, the old radio we kept on for background "machine shop" music would go into cardiac arrest every time I started an arc. We only used the TIG for the random job with stainless and for fusing together slip-fit parts. I never could weld aluminum worth a damn...
Mig has 3 modes of electical transfer, short circuit, globular and spray. You need to have 85% or higher argon gas for spray transfer. 90/10 is common. Globular is in between short circuit transfer and spray arc. Basically at the upper end of short circuit transfer but isn't commonly used because it can create a lot of spatter. Spray arc has almost no spatter but can only be used in flat and horizontal position. Some flux-core wires operate in globular transfer (1mostly spray) but not short circuit. With pulse Mig it can alternate between short circuit transfer and spray transfer. This allows it to be used in out of position welding to greatly increase production. That said short circuit Mig is used extensively for root passes on high pressure vessels and piping. The hot pass is done with stick or flux-core and I would guess penetrates into the root pass 75% or better. Really critical applications will call for Tig root passes. Some Mig welders have inductance control that can effects the pinch effect of short circuit transfer. Basically how fast the short circuits happen. One thing I've wondered is if you don't have an absolute perfect ground if stray current can go in any direction it wants until a good ground is established?? Lots of people just put the ground clamp on without much thought about it.
yeah, I'm no pro with welding, but I have a Yes welder now, and my older Lincoln AC buzzbox and a Century AC/DC buzzbox. By far the 'noisiest' with regard to AM radio is the newest IGBT Based Yes Welder. IGBT welders are two stage - they take line voltage - typically 120 or 240 VAC in the US, and convert it to low voltage high current DC. Then the inverter size converts that to a much-higher-than 60hz (or 50 for the rest of the world) frequency AC. They are great at what they do, but wow do they send EMI and RFI all over. I carry one of those meters that measures RFI, EFI and Magnetic Fields, and the buzzboxes can move the needle on it from about 5 ft away - you can play with the settings and find stray fields. But the YesWelder throws just TONS of RFI across an immense frequency spectrum. if you are familiar with how an electromagnet works - or a motor/generator for that matter, you'll realize that an electrical field that varies or moves across a conductor will generate a flow of electrons. Even if you ground really well, your advanced welder is going to push a LOT of RFI/EMI about. RFI is Radio Frequency Interferance, EMI is Electromagnetic Interferance. Strong enough fields will create noticable votages running about on your circuit boards. if you have a bunch of wiring near your weld, you are guaranteed you'll be generating power on that cabling - how much power is hard to say - EMI drops off by the inverse square of distance from source. RFI is affected differently. Disconnect your stuff, ground what you can, and don't weld close to anything sensitive. Not with today's IGBT gear. I don't know - perhaps Miller (the group that pioneered IGBT welders) has one that's not so noisy, I just can't afford that for a hobby-level welder that I am.
If there is not aq good welding ground itt doesnt become "stray current" it simply arcs. Not the same as bonded AC systems that might go hunting. Unhooking a battery "isolates" what again?
In the same way that your car never really de-energizes with the battery connected, neither do modern machines. if your welding on there with a perfect welder that doesn't generate noise, EMI, RFI, etc, then good on you! But if you don't have that kind of great welder, there's a chance you create random and stray currents with your rapidly shifting radio and magnetic fields as generated by the welder. The further you are away from the touchy stuff, the less likely you are to cause issues, yes. However, if it's powered in a stand-by state for maintaining memory/settings, etc, and you introduce random low power fluctuations, you take a chance on altering some bit or byte in the powered on ram or whatever is holding run-state. Then again, if it's all powered off, then there's no state to change, and the power required to actually do damage is significantly higher. The more wires you disconnect - eg the shorter the 'antennae' of wired going to your ECM, the less magnetically generated current you'll get. so, disconnecting the ECM guarantees minuscule currents at most. you aren't isolating stuff really - you are just breaking/interrupting the ability of EMI/RFI to be efficient at making an actual current in the wire. That's all it is - risk management.
I wanna hire a couple guys here. It's neat to be able to troubleshoot a broke welder down to someone didn't remove a couple wraps and that gremlins in these machines cause by eelding neat to learn that disconnecting shortens wires too.
If this was really a statistical risk we would serve thread a week. Once a month across all of these forums vs one a year that ends up like this. Secondly , I didn't actually tell someone not to do this. I made the point I have done this thousands of times as well as others in the biz does it thousands of times without incident. Seems there would be several threads on this , wonder where they are?
I think the machine in this thread might even have a switch.
Got one guy here burned something up ,,, twice with a welder and once with a wrench, seems we should see a few more pile on with real incidents?
What kind of beer do you guys drink? I’m gonna go grab another one. Seems like we’re going to be here for awhile. @Welder Dave @Nige @Vetech63 , I’ll grab you guys a round.
Michelob Ultra for me…….chilled to the point it’s starting to ice!
Since you didn’t offer me a beer, let me offer you this. You sir, are a legend, in your own mind. Lol Enjoy the beer, I’m a Vodka kind of guy. lol
Interesting to see where this has gone. There is absolutely no denying welding with the batteries connected has caused electrical components to fail. As to why, who cares, it does, that’s a fact. Honestly, I’ve never heard of anyone who argues that either. Best practice, disconnect batteries before welding! But as to why, sensitive electronics do not like voltage spikes. Like all the tiny microprocessors and IC chips in an ECM. They literally explode. Sberry, are you using jb weld?
I wouldnt mind seeing some of these incidents. Should be plentiful since it seems so common. Even guys arguing these points cant seem to come up with much.
Should be easy to find dozens of these then? Show me an exploded ECM.
This is not accurate. There's isn't really a thing called stray current because you need pressure (voltage) to push electrons (cause current) through a material. Electrons always exist. Some are "free" like in metals and can flow very easily with minimal voltage while other electrons are "locked up" like in insulators and require high voltage before they move or flow to conduct electricity. But trust me, with high enough voltage, you will cause locked up electrons to become free electrons. In between these two materials are the electrons in semiconductors whose ability to flow depend on their free energy, temperature, voltage potential, and a host of other things that determine how easily they can flow or conduct electricity. (There's a reason these materials are called "semi-conductors") So what you're really trying to do is minimize undesired or stray voltages to prevent current from flowing in materials. @laidback01 already answered this question very effectively as you never truly isolate any electrical circuit from RF energy. In his perfect words "you aren't isolating stuff really - you are just breaking/interrupting the ability of EMI/RFI to be efficient at making an actual current in the wire." By physically disconnecting components from easily conductive paths (just like disconnecting wires and battery cables), you're simply introducing more impedance or resistance to the circuit. More impedance means it's takes more voltage to move the electrons. Thus for disconnected circuits, any stray voltage from RFI has to be higher before it causes damage from undesired currents. As laidback01 said, it's just risk management. Don't confuse a destroyed circuit with a compromised circuit. Yes, it's easy to tell when an electrical system is affected by welding when it becomes compromised to the point of it being destroyed or useless. But not all damage caused by RFI is immediately recognized. Most damage is subtle. A voltage spike from RFI can easily cause a small copper trace line in a computer chip to heat up and recrystallize in such a way that it is more prone to metal migration. Metal migration is literally electrons knocking metal atoms into a different position and can cause these small trace lines to open and kill the circuit after continued use. That's just one example of what a voltage spike in a circuit can do as there are many, many more ways to compromise integrated chips with stray voltage. Reduced MOS-gate thresholds that change compute logic, punched gates, flipped volatile memory bits, broken compute threads, etc. are real issues and are generally caused by stray voltages. So no, just because you don't immediately kill a system with an atomic blast of RFI, with a much lower amount of RFI you can easily cause significant damage to it that overtime will cause it to die a premature death or at least behave in a way that wasn't expected. Remember the electrical gremlins Dave mentioned earlier?
No exploding takes place. It's much more subtle than that. Visually an ECM in perfect working order and one that isn't is not noticeable from the outside. The difference is that one works and the other one, as we refer to it in the industry, is "bricked". Literally meaning it's as much use as a brick. All it takes is one internal component or one of the tracks on a printed circuit board to go pop and that is all she wrote. @ahart - I'll take a glass of a nice Malbec from Mendoza instead of a beer if you don't mind.
Consider this scenario that may not have crossed the minds of some people. It must happen on plenty of remote job sites in out-of-the-way places worldwide. I just use this one as an example based on my own experience. A job site in the Amazon rain forest 60 miles from the nearest highway down a dirt road. "Express" parts delivery is probably about 14 days IF the supplier has the part in stock and IF there is space on a plane to get it into the country. Anyone who has ever worked on Cat 3500 machine engines will be familiar with the scenario. Broken exhaust manifold hold-down bolts. Welding is usually the only way to get the broken pieces out because the break is generally below the surface of the cylinder head. So you drag the welding cables 10+ feet up in the air and drape them over the machine chassis/sheet metal to get access to the top of the engine. There really is no good way to completely keep them out of contact with the machine. Yes the welding ground will be hooked right to the cylinder head where the broken bolt is located. Now this is an engine probably costing around $400k exchange, with a $4k ECM bolted to the front of it, and fitted with electronic fuel injectors probably costing about $1k each (x16). To top that off there are wiring harnesses the size of a baby's arm literally everywhere and a shed-load of electronic PWM sensors installed all over the engine. The truck it's installed in probably costs $3m and is valued at say $5k/hr in revenue loss while it is stopped. So bearing in mind all of the above then yes I am going to disconnect the batteries and unplug the two main wiring harness connectors from the ECM. The whole process will take no more than a couple of minutes. These are precautions that I consider to be reasonable and prudent under the circumstances. I prefer to do that rather having to explain why something electronic, expensive, and likely unobtainable (at least within a short time frame) got fried leaving the machine as a beached whale. This is because at multiple occasions in the past I have seen instances of machines where electrical/electronic gremlins have appeared right after welding took place. As a former work colleague of mine once opined - "electricity is invisible, and it bites."
No doubt. G series dozers have electrical gremlins on their own without adding the possibility of anymore. Just let the voltage spike while grading across a site and it locks the parking brake down... You'll be reminded of all the stuff stuck behind the seat when it lands in the floor board. Been there, done that.
Haha. Sorry, Nige, not to call you out but there is a sh*tload of exploding at the microscopic level. It's just buried inside the integrated chip (IC) that you and I can't see with the naked eye. In my previous career, I read many FA (failure analysis) reports of damaged ICs and I can tell you that voltage spikes cause mini explosions in devices that are massive and destructive when compared in scale to the device. Within just a few seconds of surfing, I found a very relevant paper on automotive MOSFET damage with pretty pics from voltage spikes. I attached it for your light reading when enjoying your Malbec (I'm a Zin guy myself). Note that paper goes into detail that these failures were not because of manufacturing defects but more than like because of another source of stray voltages introduced into the circuit. (Welding maybe? haha!) Duplicating the damage circuits in the lab required only 100v applied for 150 microseconds to cause the damage. And that's for damaging physically huge power MOSFETs which are designed for large power loads. When you think of the atomic level transistors (yes, today's transistors have active dimensions that measure less than the width of an silicon atom), any small voltage spike for an IC is the same thing as a nuclear explosion.
To clarify I was referring to external indications of exploding. The level at which most job site staff will inspect a component.
Yes, I know. But I couldn't resist the opportunity to show everyone what us 'pointy headed pencil pushers' saw that came in the field for analysis. Most people don't appreciate the substantial amount of engineering and protection that goes into building integrated circuits and also trying to harden them against nonconformal events...like close proximity welding.
All I know is, explaining to your boss why a 7k ECM fix is required, especially after hauling a road grinder a few hundred miles into Vermont, then it won’t start, to get it off the low boy, because you were too lazy to unhook the batteries for a couple of minutes of welding, gets old. Never want to do that again. Not to mention, it was a rented grinder from Milton Cat, because you fried the bearings in the drum, welding tooth bases on, because you put the ground right next to you on the leg frame. Oh, to be twenty anything again.
I’m here for every single bit of this. I’ll buy a virtual round of your choice for all of you. Except for the person that’s wrong, you caused this mess. It has led to some interesting discussion about electricity and electronics though and I appreciate that. I can’t show anything that I’ve burnt up welding because I got learned to isolate early on. I’m team isolation, who’s with me?
One thing I always appreciate about this forum is that on any given day a particular member can come up with information that I'm guessing no-one apart from those who specialize in that field has ever seen before. In this case it simply serves to reinforce the stance of those who choose to isolate before welding as a reasonable and prudent precaution.
Mosfet are a type of inverter used in earlier inverter welders. IGBT inverters have proven to be more durable and last a lot longer, at least in welder applications. Forget welding on machinery and consider if you had to gouge something to make repairs. It's almost a direct short and will make the cables jump. You can even feel the current going through the cable. I sure wouldn't have the cable laying on the machine without having everything disconnected. If you can feel the current going through the cable you can be sure it will mess with electronics. Gouging is an extreme example and welding may not be appear harmful but there's still a big spike of current, especially when first striking an arc. I would imagine most damage happens when first striking an arc. All of a sudden you're introducing 70-80 (open circuit volts) for a millisecond dropping down to around 24-30 volts but at 100 amps plus (stick welding). I don't drink since I'm taking a bunch of meds but this thread could certainly drive me to drinking. It's a case of 100 people could say to take precautions when doing something with potential for damage but there's one guy that figures it's worth tempting fate. OEM's aren't likely to post of machines that have been damaged by welding because some people could interpret it as the machine is the problem. It's much easier to post to disconnect batteries and ECM's which is based on past experience. I'm sure OEM's have documented cases of welding messing up machines. They didn't just just put the warnings out for no reason.
Most of my electronics work has been with power supplies and hi-fi class AB amps. So, I tend to work with bipolar transistors mostly for the analog items, and mosfets for switcher power supplies. My favorite is the venerable MJ3055 NPN and it's PNP brother the 2955. Now, I've looked at IGBTs and just kind of poo-pooed them as the next iteration of mosfets. They really are pin compatible in most of my use cases. Here's a bit of reading on them and their relative differences. point being - my apologies for not getting the class right for the modern welders - I'm at a hobby level here. I just don't use them they way that you folks do.
You didn't get it wrong. Your input is very valuable. Mosfet were really common in early inverter welders. I just remember reading about different kinds of inverter welders and discussions between Mosfet and IGBT. I know that if something goes wrong with an inverter welder it is often cost prohibitive to repair them. Know a shop with an early Miller (Powcon) XMT300 and it was $3000 for a new board and this was probably 20 years ago. Same thing with the Miller inverter Tig welders. Once they die they're pretty much done.
>I know that if something goes wrong with an inverter welder it is often cost prohibitive to repair them. Know a shop with an early Miller (Powcon) XMT300 and it was $3000 for a new board and this was probably 20 years ago. you know the funny part here is the amount of advancement we've had in just 20 years. CanBus has invaded heavy machinery to various degrees based on size and make. Most of the modern systems encrypt their on-machine communications - literally the communications from the transmission to the ECM, the engine to the ECM, and back again, it's all encrypted. the complexity in your IGBT welder is absolutely dwarfed by the full computing systems which are running in modern equipment. Hell, the average ag tractor (2020+) has more compute power than a datacenter in the early 90s did... they have GPS, obstacle avoidance, self driving, self planting, and I guess some handle harvest now. it's amazing (and scary) what these machines can do. the point to this essay is this: these electronic systems are really durable and tough to a LOT of things. But they are fragile when it comes to certain factors. We have the ability to limit that risk through simple actions. it brings up another thought - I had a boss who couldn't get near a compute or a phone for more than 30 - 40 minutes at a time, or the device would bluescreen or crash. My wife can't walk around this house without shocking herself on every metal surface she finds. I, on the other hand, have one hell of a time building a static charge and have to kind of work at it. Some people are touchy to celltowers and get sick - they don't bother me in the least. So, it goes further, and there might be welders who have a natural field about them that quells the problems of stray RFI/EMI, and others who don't have this natural aura. I don't know what it is, why it is, but it's observable. Some people have really distinct electrical traits...
Modern stuff gets so many gremlins , as proven by multiple posts how could one ever identify the difference.
Hey @rockbottom , Did we ever answer your original question? Hope you come back.
I'll play devil's advocate here. I have welded on numerous machines over the years and have never lost an electrical component. You guys can spank me now.
Nah, no spanking needed if it was your own equipment. But if it was somebody else's equipment and you didn't follow precautions, well, maybe a spanking or at least making you drink a hot, flat beer is warranted. Isolating electrical components when welding is like wearing a seatbelt when driving. Just because I don't wear a seatbelt (or isolate) doesn't mean I will get into an accident (or kill my ECM). It just means I'm increasing risk to myself to go from a situation where I probably wouldn't get hurt to now getting a minor injury or worst case sustaining a fatal injury. Most integrated circuits have numerous built in protection features that protect against unwanted voltages, noise, and such. The biggest one is Electro-Static Discharge (ESD). That's the little static spark @laidback01 was referring when someone gets zapped by the door knob after walking across the carpet. ESD from the human body can easily exceed 10,000 volts but the charge is pretty small so the protection in most circuits can shunt it to ground if needed. But again, that is "most" of the times it works and it's why you handle all electrical doo-dads with care. Lightening strikes are another hazard that's taken into account for circuit design. Strikes in the general vicinity produce a huge EMI and most of the time, the circuit is protected. But if the strike is too close, the EMI overpowers the circuit and out goes the ghost. As Dave said earlier, "I would imagine most damage happens when first striking an arc." Bingo! He's spot on. That initial arc acts like an old timey spark gap radio (massive EMI) which are illegal now because they emit so much noise and cause interference in circuits. Welding on your equipment is like intentionally making mini-lightening strikes on your machine. Why wouldn't you do your best to protect the expensive electrical doo-dads if it only takes a couple mins? I'll change up the conversation a bit: What do ocean tides and EMI/RFI in circuits have in common? In select areas, the peak amplitude can be much higher than the overall average amplitude because the energy gets channeled or condensed into a small area due to the geography (or circuit). We can predict ocean tides pretty well because we measured them over thousands of years. Trying to predict circuit spikes because some random form of EMI gets coupled with some random inductance factor on some random piece of equipment getting channeled into some circuit is too complicated. Instead engineers set very high limits of electrical protection and call it "good enough." If and when you ever exceed that limit by welding without precautions is up to you.
Lost a nice watch once. Stopped it dead and a new battery only made it turn backwards for about 30 seconds. Usually don’t wear a watch anymore.
Funny thing is I will kill a watch battery in 1/2 it's lifetime so I don't wear one anymore. Have a kinetic one but it stopped working as well. My BIL has the same problem. Weird.
The aliens are among us!
When I first got in the military, it was beat in me to wear a watch, and it had to have the luminescent dial and hands for obvious reasons. As I progressed in my career and started working on aircraft electronics, it was beat in me to not wear a watch for safety reasons. No jewelry on the flight line nor especially in the backshops on the electronic repair benches. Then I started being a crew member and it was beat in me to wear a watch again but only when flying. When on the ground doing maintenance or pre- and post-flights the watch had to come off again. Now I have a smart phone that tells me the time in big letters so I don't have to put on my readers. Somewhere in a box somewhere in the house I have my old green Timex with the velcro cloth band. I wonder if the dial and hands are still glowing.
I thought this thread had run it's course, guess not. Seems everyone working on equipment works on a cat cost a million dollars. Wonder how much damage is done every year fughin with batteries? We should be seeing regular threads here with damage. Even the guys all of a sudden become electric experts can't seem to come up with a actual damage cases. Seems there would be a few. If it followed the speculation here it should be in the hundreds, I should have done it dozens of times.
I believe this has been looked at by engineering s.nd it usually goes about like this thread. The end result was a couple of them simply said. We will just put this in the manual and be done with the debate and it's not skin off out AZZ,, next guy says,, well Cat did it so it must be so, we will do it. Now when it comes to this the thread parrots jump on it. Did you know that if you don't change oil in your lawn mower every 6 months. Used or not it will blow up,, says so in BMW ownerssnusl. Must be so. Everyone writes it in the manual, anyone think this is accurate? Often repeated, if you install an outlet without inspection your insurance won't pay in a fire. This is repeated thousands of times , never seen it happen. If you don't drain air comp tank every use it will blow up,, they been passing the SAME picture on those threads for 20 years. Not saying it has never happened but I don't think the risk matches the speculation,, back to the fact we don't see threads from a dozen forums.
We can't even get 2 posters in the same thread with actual damage. All what if,, what would you tell your boss kind of posts, not a thing to do with actual fact or events.
Had a similar thread on .SFT about 20 years ago. About 100 posts to same type of question, was an old tractor maybe and 1 guy comes on, some sprout says he struck arc on his S10 and blew all the lights out, ruined it all. A long list and says he literally exploded the lamps. Out of 100 only a couple said bullsnit and it carried on with same as here. All opinion, few facts. Same as a couple guys here that have realistic view. They simply give up and the belief is way stronger than the reality.
heh, good thing is we are all entitled to our opinions. It's a generally easy going forum, and even if you don't get others to see 'it' your way, the outcome matters little. Do what works for you! I'm going to keep disconnecting the batteries on my equipment and vehicles when do any welding. it's what I was taught, it's what I'll keep doing. The time used is minimal, and so far so good! have a good day, do your own thing and enjoy life! I'm going to go and fix the frogs (shanks) on my excavator, I need to put some more padding in place under some of the teeth to keep them firmly in place. they rattle right now. Yes... I'll disconnect the battery before starting this simple welding job.
Yeah, I get it. I might as well take my toaster into the shower too since there are no documented deaths of electrocution from taking a bath with a toaster. Google link to find deaths attributed to toasters in the tub While deaths from hair dryers and heaters from falling into the tub are actually well documented cases of electrocution, making toast while soaking in the tub does not appear to be a hazard for electrocution because I just can't find it being documented in real life cases...just like welding does not appear to be a hazard for damaging voltage spikes in electrical circuits.
. Typical of this type of thread. No one said not to do this if you feel like it, great, makes a guy feel more handsome go for it. But now we are trying to connect toasters to this logic?
@Sberry - I dunno what anyone else thinks, but I reckon you could probvably start a fight with yourself in an empty room.
It's fairly easy to see the toaster logic and it is supported by actual events. More than 2 and they come up with GFCI.
I got a lot of respect for your tech aptitude and experience, don't ruin my image by becoming a MOD EDIT along with the world's greatest welder Watch the language.
I left this thread die once before but guess I ain't the final authority. It was a simple question to start with,, should be easy if this is such a fact for someone with 50 years to come up with a case or 2,,, we got 1 guy ruined a backhoe and a motorcycle and hurt himself with a wrench and a wrist watch. Should be some real evidence elsewhere, somewhere? Should be a dozen events posted to this thread if that was really accurate.
This thread - Think it's time to move on....