I have been reading this thread, and have some thoughts and info that may help when making choices about forges and temperature measurement.
If you drive a Yugo, it won't matter how fancy the Michelin tires are, it still won't be a high performance car.
In the same way, if you have a soup can forge, you won't get a highly accurate HT, or much of anything quality wise. Using a pyrometer will help a bit, but it won't compensate for the extreme temperature variations in such a forge.
If I could recommend any single expenditure as being money well spent, I think it would be making a proper forge. This might cost all or $100 at the top end, and as low as $30 with some scrounging skills, but will save you hours of work wasted on a bad HT or a poorly forged blade. An 6-8" diameter by 12-16" long piece of pipe and some refractory supplies from High temp Tools and Refractory will make a very serviceable forge. A bean can lined with Plaster of Paris is just not going to work nearly as well. Propane burners can be made from scrap piping and a few things from the local hardware store for almost nothing,...... using a $30-50 propane plumbers torch is a waste of time as well as a waste of money. Add a cheap ebay PID and TC, and you can tell what range the forge is running at. NOTE - The TC is not necessarily telling you how hot the blade steel is!!!, that is why they are almost useless in a small "can" forge with minimal refractory and insulation. In a regular size forge, after the forge has run for a while and soaked the refractory to even out the chamber heat, the TC can be a fair indicator of the blade temp.
For a little more money later on, you can make the forge run automatically with temperature control by making it a blown forge with a two-stage controller.
The sale price of one hunting/fishing knife can pay for the whole forge, and the sale of a second one can pay for the PID controlled burner setup.
For checking a blade's temperature in a forge with poor heat control and limited space ( or any forge/kiln), probably the best and cheapest method is to use Tempil-Stix. You don't need a big set of them when using these type forges. A set of 1400F, 1450F and 1500F would do just fine. Mark the blade in several areas to determine how even the heating is going.
http://www.tempil.com/products/tempstik-test-kits/
Another method is the pyramids and cones called Segar cones. They slump down when they reach their rated temperature. Just like the crayons, a set of three temps would help you a lot.
Getting a low cost PID and TC from ebay is good for a low budget starter system, but a better grade PID controller unit, a 10 gauge TC with ceramic spacers, and a 3/4" ceramic sheath from Auberins or another quality supplier is money spent once. If you know your stuff about PDs and TCs, you can find what you need on ebay, but sadly, most people look at the price and stop there. This is one of those places where cheaper usually does mean less quality. The types of TCs and stuff you need is in the Sticky on PID Control and forges. All TCs are not the same, and there is more to it than just hooking stuff up with wires.
The info is elsewhere, but the basics are;
A TC needs shielding from the erosive atmosphere in the forge. The cheap stainless/monel clad TCs will hold up for a while, but will burn up eventually. They also usually have fairly thin bi-metal wires. What you want is a 10 gauge (1/8" thick leads) type K thermocouple with 12" long leads. It should come with ceramic spacer beads to keep the leads apart. You then need a 3/4" ceramic TC shielding tube. 8-9" is plenty long here. You will need a TC terminal block to attach to the end of the TC, and several feet of type K connector wire ( attached to the TC connector block, and at the PID). The wire and the TC are polarized, so make sure you hook + to + and - to -.
The PID can be of several types, but make sure it runs on the voltages you use, and takes type K input. In a PID, you usually get what you pay for. Spending $15-20 more for one from a know source is probably a good idea for the novice.
