cellulosic -- DC-ve (DCEN) or DC+ve (DCEP)

I haven't done a downhill root pipe test myself so I can't say if you do run electrode negative. I did a plate test years ago that called for roots done with 3/32 E 7018 in all positions. The root was done electrode negative and then the fills and caps were electrode positive which is normal. If everything is going well you often can just rest your electrode on the lands and most of the arc is not visible on the operator side. That of course is with a perfect preparation unlike the real world. I have always done roots with 6010 electrode positive. I just did what I was told to do. Experimenting around in college is a good thing to do. Once you are involved with a job time is money and the philosophy is: " It worked last time so lets not change it." Roots run on massive thicknesses behave very differently than when the material is only 3/8 to half inch thick. The large mass seems to suck the heat out of your weld pool even with preheat. Randy

Pipe-welding with cellulosics (xx10's) - I simulated this by standing-up vertically some V-prepped plates and welding downhill. Got keyholing and full penetration, running fast with "scratching" contact with sides of V-prep.

I connected DCEN, because I heard that is what you do with cellulosics. But I look at eg. Lincoln info. for "Fleetweld" and I see "DCEP" as the recommended. Confused :-(

Technically, connecting DCEN gives more penetration? You get 2/3rds of heat liberated at the +ve pole, so you will get more heat in joint / less metal deposition rate DCEN compared to DCEP? Root-running DCEN would make sense if you can have a thick root which is less delicate as you swing around long pipe-lengths with a crane, plus the weld will be fast-freezing, having a deep full-penetration narrow weld? So you can root-run fast downhill with the metal being held in place by surface tension to a lot of (thickness of) surrounding solid metal?

Technically,no. For a non-consumable, bare, electrode as used in TIG, DCEN provides more heat into the weldment than does DCEP. When you start talking about SMAW the issues become a little different. In SMAW, no one uses bare electrodes anymore. They all have a flux covering which in addition to other benefits, can alter the arc characteristics to the extent that for the 6010, DCEP has the greatest penetration and the least amount of deposition, compared to DCEN. Try a little test: Scrape off all of the flux covering for (2) 6010s. Run one of them at DCEN and the other at DCEP. I think you'll find that the DCEN has better penetration. In other words, the flux covering has altered the arc characteristics.

-dseman

You may also want to check out the following from Lincoln on pipeline welding procedures:

-dseman

The majority of our work is downhill pipeline related welding and here is my view. It sounds like you are not in the U.S. and maybe in the U.K. Is that right? In the U.S. beads (the common name for a root pass in downhill pipeline welding) are run with the electrode positive in almost all cases. I hear that in other countries electrode negative is common. Also in the U.S. beads are almost always run with Lincoln brand 5P+ rod, sometimes 70+ . When welding the bead, the rod is in contact with the bevels of the pipe all the time, pressure and angle and speed are all constantly manipulated to compensate for variations in land size and space and hi low. Electrode negative results in less penetration and lessend chance of burnthru. Keep in mind that on a pipeline job, at least in the U.S., you will have no choice in what polarity to use or what type or size of rod to use for each pass. All of these things are dictated by the procedure in use.

JTMcC.

Yes - that's absolutely correct. I am in the UK

For the issue of the DCEP vs. DCEN of cellulosic electrodes on the root pass, the "normal" use is DCEP, which gives much better penetration. However, DCEN can and has been used successfully (primarily outside the US). There are some applications where excess internal root reinforcement is undesireable. Use of DCEN helps minimize the internal root profile.

For the issue of heat balance and penetration, DCEP gives more heat to the base metal, thus more penetration. This is true for SMAW ("stick") welding and GMAW. The opposite is true for GTAW - DCEN gives more heat to the base metal. This has to do with the "work function" (i.e. high-falootin physics) of the tungsten. Which also leads you to why AC, rather than DCEP, is used for GTAW welding of aluminum. Anyone (like me) that has gotten the polarity wrong when running GTAW will quickly discover the heat balance effect after they have melted their tungsten electrode.

Note that because of the heat balance ratio for SMAW, the burn-off rate for DCEN will be significantly higher than DCEP. This typically means that the current setting would need to be reduced when compared to previously used DCEP current settings.

I can provide additional information and data if anyone needs it (although it might take a while to dig it out of my old files).