If you're just getting into this, things like junkyard steel and unmarked imported steel are out. You may have some problems on your own, and having steel that perhaps wasn't processed correctly or is not evenly alloyed across rolled sheets is just a no go. I have had this happen several times, though one of the flawed suppliers provided 115crv3 (not that important, just referred to as Silver Steel Rod) that was both lacking in uniformity when snapping samples, and unhardenable - even in brine. Something was wrong with it, I don't know what, and the seller took it back and provided a refund. I have pictures of the internals, but will spare you - the snapped samples may not mean that much.
Beyond that, both tested plane irons at very low price points and other samples of imported (non-Europe) steels of low cost alloys have shown problems that are big enough to show up on wood. We'd like to be far from that.
You can practice on high carbon springs or something from a junkyard, but I'd be hesitant to use anything like that for a good tool. The savings is too little for the time involved.
My sources for various alloys, all of these good quality are:
* 80crv2 - New Jersey Steel Baron (steel from Buderus). Others are probably fine, but price varies widely and this is an inexpensive steel. If it's expensive somewhere, why bother.
* 1084 - New Jersey Steel Baron (steel from Buderus)
* 1095 - New Jersey Steel Baron (steel from Buderus)
* 26c3 - Alpha Knife Steel (Voestalpine/Bohler is the source). 125cr1 from New Jersey Steel Baron is an alternative and much lower in cost, but samples I've broken are less uniform.
* O1 - McMaster Carr (Starrett steel locally), Alpha Knife Supply (Bohler), New Jersey Steel Baron (Buderus)
* 52100 flat bar (I think you should avoid 52100 unless you've mastered other steels - it's more difficult to get good results with) - New Jersey Steel Baron (Buderus).
* W1 (McMaster Carr or Die Supplies - steel is Precision Industries) and W2 (New Jersey Steel Baron, Buderus again). W1 is a one-off here - Precision provides good quality bar in a tight range with 1% carbon, but the spec is huge and unknown product may not be worth getting)
* Hitachi White - available from New Jersey Steel Baron, very expensive - 26c3 is a better choice in an almost identical alloy unless you want bragging rights.
There is a theme here - nothing above needs a complex heat treatment regimen, but 52100 is more difficult to get to full hardness and it's not very good for woodworking at lower hardness due to high toughness and desire to roll and hold on to a roll. 52100 is described as a steel that needs an electric furnace to be right, but this isn't really true. The nuance is the alloy loves to hold the carbon in carbides and leave too little in the matrix for an easy high hardness result. You have to figure out a process that won't grow the grain, but that will get some of the carbon out of the carbides into the matrix. Some steels really have too much carbon in the matrix (O1 and 1095, for example) to ever have high toughness, or resistance to breaking by bending fast or slow, and others like 52100 and 80crv2 to some extent want to hold on to carbon in their carbides and make you work to get it out. I no longer find either of them difficult, but it took off and on trial of samples to get a 52100 sample in the upper or above the upper hardness spec from the heat treatment sheets. It's not really for beginners, and the structure that it's often delivered in - spheroidized where a long slow below-nonmagnetic heat allows dissolution of more and more of the carbon into big round carbides makes it a test, even for electric furnace schedules.
These steels will be explained in another post along with what I think is the best way to deal with heat treating them. It varies little, and you can figure it out on your own, but you can start from my advice and be further along.
Knife forgers are always encouraged to use 1084 first, and I think that's bad advice for woodworking tools - O1 is a much better starting point.
Notice, there is no 5160 here, which is common on forged in fire and other shows, and common for "knife people" because it's tough. I've observed Larrin Thomas's discussions long enough to realize the two biggest issues with knives are lack of toughness (broken knives will be complained about and returned long after the warranty is over) and stainless knives that don't avoid all stains. Consumers don't really understand that stainless is a subjective term and not absolute. Quite often, the stainless steels that make great cutting knives and also will make a good plane iron just don't do well for uneducated consumers, and the hardness level is above steeling.
Forged in fire is an interesting show, but what you can really use it for is once you have some experience, identifying the skilled workers on the show vs others, and there have been some legitimate really fine makers on there mixed in among people who have obviously never snapped a heat treated sample.
Lastly - a good retailer will link the melt sheet from your sample. Alpha Knife Supply doesn't do that, but rather writes a code on your bar stock and is protective about some things, and others, not. For example, Bohler O1, and then next, 1095 of unknown origin, which I found to not be worth buying. I have never ordered a steel sample with a specified mill and melt sheet and had an issue with it. it's rare.
This is a Buderus melt sheet from NJSB. You can clearly see the spec range, that the steel came from Buderus, comments on manufacture (electric process and ingot casting - as in, not powder or sprayform or anything exotic), and sometimes these sheets will describe a percent spheroidized. This one does not, but it's 1084 and spheroidizing probably doesn't do much because the amount of excess carbon is small. You can read about things like eutectoid limits elsewhere, but the sample here describes 0.81% carbon and the steel has no chromium that would express an interest to suck in some of the carbon.
The list of retailers here is who I have gotten steel from. I've also ordered from Carpenter and USA Knife Maker (the steel was something obsolete in one case, and the same 1095 that NJSB sells in another). At some point, you're not going to be a customer of every single mill because you actually have to make things with the steel. Bar stock is a crappy collectible, especially if it's not stainless.
it is possible to prove my suggestion wrong in cases. As in, you could find superb 1084 with no known origin and then confirm that the bar came from a specific melt and get a whole bunch. I think that kind of thing is better left to manufacturers who may get some kind of financial realization for saving $1.50 on a tool or something that they're making thousands of. If you think it's hard to use bad wood woodworking, try to find something you want to make out of crappy steel.
Added later 44 min 11 s:
One extra comment here - you will find steel that is ground, or not ground. And within grind types, NJSB, especially, offers wide belt sanding of steels in place of grinding.
The grinding marks from a wide belt sander are so deep that I'd avoid it. It's more work for a small shop user to deal with removing the grind marks than it is to not have the steel ground, and the wide belt grinding comes with an extra charge. Otherwise, ground or not is really up to you.
I would assume the process of wide belt sanding is better for knife makers who are going to do significant grinding of the bar in the first place, but with some flat tools like plane irons, you're not going to want to deal with it. Ground material generally has more material ground from the surface than I've found necessary to avoid reduced carbon in the surface layer ("decarb")