[LearningFast]

Greetings,
I have occasion to extract a stuck 6” long pushrod from a hydraulic brake system. It has been suggested to weld a smallish piece of steel to the offending, stuck pushrod, utilize this piece to pull/ manipulate the rod until it is removed. Object’s surface is flat, circular, pointing up, and between 1/4” and 3/8” diameter. Composition is steel, unknown alloy.
My concerns are many.
1) how local can the current be kept? I’m thinking any current that flows between small gaps could arc and produce undesired results in places other than the weld location.
2) is it possible to bring a 90 degree local ground needle in from the side, to the actual object I wish to weld to?

If this object cannot be welded to, the extraction is much more involved. Carbs, intake manifold, tapper cover must be removed to access offending part.
It would be super nice to get it out from the surface….somehow….

Inputs appreciated.

[tweake]

you would need to get the earth onto the pushrod, otherwise it will arc off the rod onto something else. make sure the earth is clamped down well.

[cj737]

The pushrod has a path to ground anywhere. It touches a rocker shaft, a cam lobe, and in turn the engine block. The block is mounted to the frame via the mounts, and likely a battery ground to the starter. So don’t worry about “locally” grounding the pushrod.

To weld to it, you need to chemically clean that pushrod end as well as you can. (Non-chlorinated brake cleaner works well). Then weld to it with MIG or TIG. With MIG you’re likely to get some spatter, so bear that in mind.

My fear would be the pushrod is a chrome moly steel and thus you’d want to TIG weld it since MIG and ChroMo don’t like each other too much.

Clamp your ground to the cylinder head and get to it.

(Your title states engine manifold, but your description says brake cylinder. So clamp to the nearest body or engine piece).

[Jack Ryan]

The pushrod has a path to ground anywhere. It touches a rocker shaft, a cam lobe, and in turn the engine block. The block is mounted to the frame via the mounts, and likely a battery ground to the starter. So don’t worry about “locally” grounding the pushrod.

To weld to it, you need to chemically clean that pushrod end as well as you can. (Non-chlorinated brake cleaner works well). Then weld to it with MIG or TIG. With MIG you’re likely to get some spatter, so bear that in mind.

My fear would be the pushrod is a chrome moly steel and thus you’d want to TIG weld it since MIG and ChroMo don’t like each other too much.

Clamp your ground to the cylinder head and get to it.

(Your title states engine manifold, but your description says brake cylinder. So clamp to the nearest body or engine piece).

CJ must be much luckier than I am. If I attempted to weld a pushrod (of any sort) with the return clamp connected to the block, every connecting piece in the current loop, including followers and bearings, would be welded together. You might clean the intended weld area, but all the other "joints" in the current loop will be dirty and arc.

I agree that the pushrod material is likely to be challenging to weld.

You've gotta ask yourself one question: "Do I feel lucky?" Well, do ya? (Apologies to Harry).

Jack

[Hammaa2]

Greetings,
I have occasion to extract a stuck 6” long pushrod from a hydraulic brake system. It has been suggested to weld a smallish piece of steel to the offending, stuck pushrod, utilize this piece to pull/ manipulate the rod until it is removed. Object’s surface is flat, circular, pointing up, and between 1/4” and 3/8” diameter. Composition is steel, unknown alloy.
My concerns are many.
1) how local can the current be kept? I’m thinking any current that flows between small gaps could arc and produce undesired results in places other than the weld location.
2) is it possible to bring a 90 degree local ground needle in from the side, to the actual object I wish to weld to?

If this object cannot be welded to, the extraction is much more involved. Carbs, intake manifold, tapper cover must be removed to access offending part.
It would be super nice to get it out from the surface….somehow….

Inputs appreciated.

If welding is an option, keeping the current as local as possible is key using a low setting and a precise ground connection should help. A 90-degree needle ground might work, but make sure it has solid contact. There’s definitely a risk of arcing in gaps, so precision is everything. If it doesn’t work, you’re in for some teardown… but hopefully, you can avoid that

[tweake]

i would be worried it would arc inside a master cylinder, or in a linkage. places you never want welds or arc marks. hence why i think ground on the rod itself is a safer choice.

[cj737]

CJ must be much luckier than I am. If I attempted to weld a pushrod (of any sort) with the return clamp connected to the block, every connecting piece in the current loop, including followers and bearings, would be welded together. You might clean the intended weld area, but all the other "joints" in the current loop will be dirty and arc.

Electrons don’t run amok throughout the engine. Follow the reasoning here:
If this is indeed an engine, a pushrod touches a lifter which sits upon a cam lobe. That cam lobe is along a shaft that has adjacent to it lobes either side. Those lobes are in contact with lifters, then pushrods, then rocker shafts, back to a cylinder head.

So… if you connect to the head, the path between your arc (pushrod patient 0) the electrons will follow the path to the nearest ground (next lobe and back to head). It’s really quite safe.

Now, if you were welding on the cam shaft directly, and clamped to the block with no lifters or pushrods involved, then definitely passing through a bearing is very likely and I would not do it either. I’d clamp directly to the cam shaft in this example.

Electricity is very easy to manage as it has definable properties and paths.

But I remain confused because he says engine manifold then states brake piston. So if it is indeed a caliper or hydraulic pump, interior seals could be an issue from any passive heat. Until the OP clarifies his actual circumstance it’s a lot of supposition.

[Jack Ryan]

CJ must be much luckier than I am. If I attempted to weld a pushrod (of any sort) with the return clamp connected to the block, every connecting piece in the current loop, including followers and bearings, would be welded together. You might clean the intended weld area, but all the other "joints" in the current loop will be dirty and arc.

Electrons don’t run amok throughout the engine. Follow the reasoning here:
If this is indeed an engine, a pushrod touches a lifter which sits upon a cam lobe. That cam lobe is along a shaft that has adjacent to it lobes either side. Those lobes are in contact with lifters, then pushrods, then rocker shafts, back to a cylinder head.

So… if you connect to the head, the path between your arc (pushrod patient 0) the electrons will follow the path to the nearest ground (next lobe and back to head). It’s really quite safe.

Now, if you were welding on the cam shaft directly, and clamped to the block with no lifters or pushrods involved, then definitely passing through a bearing is very likely and I would not do it either. I’d clamp directly to the cam shaft in this example.

Electricity is very easy to manage as it has definable properties and paths.

It welding a pushrod in an engine, the welding current path is:

Block -> journals -> cam shaft -> follower -> pushrod.

Each instance of "->" is a place where you don't want excess heat, arcing or, in the worst case, welding.

If this is a pushrod in a brake master cylinder, the current path is:

Cylinder housing -> cylinder -> pushrod.

Same problem.

But I remain confused because he says engine manifold then states brake piston. So if it is indeed a caliper or hydraulic pump, interior seals could be an issue from any passive heat. Until the OP clarifies his actual circumstance it’s a lot of supposition.

Yes, me too.

Jack

[cj737]

If welding a pushrod in an engine, the welding current path is:

Block -> journals -> cam shaft -> follower -> pushrod.

We disagree. The head is bolted to the block. The pushrod extends from rocker arm (which is moved by the pushrod) to the cam lifter. The lifter sits atop a lobe. Journals typically divide the camshaft in 4ths, so any pushrod has another adjacent to it for the path to return to the head (clamp to the head). The block need not be the ground point, and cam journal bearings aren’t normally in the path.

I’ve welded to broken head studs(numerous times) for extraction and NEVER had an issue with journal or bearing damage. I have not welded a pushrod in an engine or a brake system, so I can’t attest to the weldability of that item.

[Jack Ryan]

If welding a pushrod in an engine, the welding current path is:

Block -> journals -> cam shaft -> follower -> pushrod.

We disagree. The head is bolted to the block. The pushrod extends from rocker arm (which is moved by the pushrod) to the cam lifter. The lifter sits atop a lobe. Journals typically divide the camshaft in 4ths, so any pushrod has another adjacent to it for the path to return to the head (clamp to the head). The block need not be the ground point, and cam journal bearings aren’t normally in the path.

I’ve welded to broken head studs(numerous times) for extraction and NEVER had an issue with journal or bearing damage. I have not welded a pushrod in an engine or a brake system, so I can’t attest to the weldability of that item.

I guess we'll just have to disagree but I'll make two points:

1. there is always a bearing or follower between the block/head and the pushrod that should not pass welding current.
2. there are no bearings or follower between the block and a head bolt - it is safe to weld.

Jack

[cj737]

1. there is always a bearing or follower between the block/head and the pushrod that should not pass welding current.
2. there are no bearings or follower between the block and a head bolt - it is safe to weld.

1. I’m curious, what engine are you thinking of that exemplifies #1? I know of no engine that has a bearing between the block and the head. A gasket, sure. A bearing, no. And cam journals don’t occur either side of every lobe pair.

2. It is true there is no bearing surrounding a head bolt, but to your point, bearings are in the block and thus are susceptible to current passage. Or do I misunderstand? It seems this point enforces mine.

I’m not arguing with you at all, Jack, these posts are tone-deaf certainly. But if you were to weld to the pushrod, exactly how much heat do you think is really going to be transmitted from the tip of the pushrod (say 110-180mm in length) to a bearing that’s separated by a lifter, a cam lobe, the camshaft, etc?

And the amount of electrical current that “might” pass by that bearing is so minimal and so short in duration, that the probability of damage is minute. It’s really no different than the voltage running between the starter motor and the block. Engines see electrical current throughout them while they’re running. All of them.

This whole caution is a bit mythical. I’ve welded so many things while the metal is in the chuck on my lathe. And the spindle bearings have never been damaged. I simply place the ground clamp between my arc and the bearings. Path complete. I think someone many years ago got this idea in their head, and we’ve stopped analyzing the actual facts and merely accepted the “never do this” approach. I’m not at all recommending throwing caution to the wind, but if in fact the OP can weld to the pushrod (remember, THAT is the real question) then I can’t envision any harm due to electrical current.

Perhaps I’m wrong. Either way, I appreciate the feedback from your end!

[Jack Ryan]

Assuming a valve pushrod for the moment.

Here is a rough diagram showing four places (red circles) where welding current should not flow, but is in the current path.

Welding a Pushrod.jpg (59.83 KiB) Viewed 1530 times

Bearings that are not in the current path are not in danger of damage.

(I'll probably need two shots at attaching the image as I get a failure on every post.)

Jack

[Jack Ryan]

Welding a head bolt or other stud is different. There are bearings within the block, but not in the welding current path

Welding a Head Bolt.jpg (36.58 KiB) Viewed 1527 times

Regards
Jack

[cj737]

Thanks, but your posts did nothing to inform me of things I don’t already know, and, don’t dispel my point.

To each their own I guess.

[Jack Ryan]

Thanks, but your posts did nothing to inform me of things I don’t already know, and, don’t dispel my point.

To each their own I guess.

Yes, of course, but I don't think we are on the same track (or page as you might say).

I am not suggesting that the transmission of heat might damage bearings, it is the welding current through a resistive contact (the bearing) that may arc and, in extreme cases, weld themselves together.

Bearings are in the block but they are not always in the welding current path. In the case of welding head bolts with the return connected to the block, for example, there will be approximately zero current running through any bearings. In the case of the pushrod, all of the welding current must pass through one or more bearings so the probability of damage is very much greater.

It is possible to weld through bearings for some time - if you are lucky - but there will come a time when a bearing will be damaged and either welded together at the time, or seized later as a result of damage to the surface of the bearing.

A similar situation occurs when welding near vehicle electronics like ECU/MCUs. You can be lucky for a long time, but unless you are very careful, sooner or later you will misjudge the welding current path. Welding on a tray (back of the truck used as a welding table) is safe as the welding current is confined to the tray.

Personally, I won't weld through bearings, but all I am advising is caution. I would guess that the manufacturer's guidelines would prohibit welding through bearings.

Anyway, those are my thoughts, yours differ - that is life.

Regards
Jack

[cj737]

In the case of the pushrod, all of the welding current must pass through one or more bearings so the probability of damage is very much greater.

This is simply not true. There are ample situations where a pushrod current passes thru one rod, down the camshaft, then back up again another pushrod without traveling through a bearing. I think this is where we differ the most. The most significant point is: not even sure you can weld a pushrod! Current issues be damned

A similar situation occurs when welding near vehicle electronics like ECU/MCUs. You can be lucky for a long time, but unless you are very careful, sooner or later you will misjudge the welding current path. Welding on a tray (back of the truck used as a welding table) is safe as the welding current is confined to the tray.

Here again, I’ve welded on so many vehicles and have never disconnected a battery, damaged an ECU or PCM, not even damaged a light bulb. It’s all about where you place your ground and how localized your arc might be to these components.

A bit of anecdote about myth vs reality (not related to welding but sort of relevant)… I know so many people who ride motorcycles. More than 90% of them will repeat the myth about not jumping a motorcycle dead battery with a car, or with a donor vehicle motor running. They somehow “think” the capacity of the car battery will damage the smaller bike battery. They know nothing about capacity vs draw, yet they repeat these urban myths. So I guess where welding on a vehicle comes into play, I tend to be more practical than stupefied or mystified. Might be wrong, but time has not dealt me that card (yet).

Hope the weather is not too hot down under, Jack nor bush fires flaring up in your area.

[cj737]

Jack - here’s a scenario you need to think through that might help dissuade your caution.

Think about a rotary positioner. You clamp to the ground bung, secure the work in the chuck, then weld your butt off full speed, for multiple passes. That machine has an electric motor and bearing and bushings through the spindle to rotate the chuck. Why then aren’t they damaged by the current?

Certainly this is a relevant example of being able to safely weld where bearings and bushings are present.

[cj737]

Jack - here’s a scenario you need to think through that might help dissuade your caution.

Think about a rotary positioner. You clamp to the ground bung, secure the work in the chuck, then weld your butt off full speed, for multiple passes. That machine has an electric motor and bearing and bushings through the spindle to rotate the chuck. Why then aren’t they damaged by the current?

Certainly this is a relevant example of being able to safely weld where bearings and bushings are present.

[Jack Ryan]

Jack - here’s a scenario you need to think through that might help dissuade your caution.

Think about a rotary positioner. You clamp to the ground bung, secure the work in the chuck, then weld your butt off full speed, for multiple passes. That machine has an electric motor and bearing and bushings through the spindle to rotate the chuck. Why then aren’t they damaged by the current?

Certainly this is a relevant example of being able to safely weld where bearings and bushings are present.

A decent positioner would have slip rings or the equivalent to bypass the bearings. At the very least, they are designed for the purpose.

Most industries warn against passing welding currents through bearings and most have written procedures to mitigate damage:

• https://www.airvision.be/en/23-electric ... e-bearings

• https://cdn.skfmediahub.skf.com/api/pub ... medium.pdf

• https://engine.od.ua/marine-engineers/u ... -Guide.pdf

Greg had a shot of demonstrating the problem with limited success

• https://www.youtube.com/watch?v=ysVIF02eiYg

Circumstances vary a lot - sometimes only a small amount of current can cause significant damage, another time a large current might cause no visible damage.

Anyway, we disagree. Let's leave it there.

Regards
Jack