Extech manufactures all of the Craftsman line of digital multimeters.
In my personal opinion, they also manufactured the Craftsman DMM that is alleged to be a Fluke 17b. (or was it 17a?)
My theory is that somebody at Fluke decided to see if they could have an acceptable quality entry level feature-rich DMM manufactured in china, and comissioned Extech (aka Topman Industrial) to manufacture a model to Fluke specifications.
Edit: Forgot to mention, Made-in-USA Fluke meters have Fluke ASICs at the heart. Extech and "Fluke 17b" have the same FTDI ASIC, and similar board layouts.
If they had totally failed to meet expectations, this would explain why this fluke model was so short lived and is so rare. And why they don't mention it anywhere on the fluke website, but google can dig up a pdf with calibration instructions for it.
Having seen pictures of the inside of the model in question, and having personally inspected the interior of another craftsman/extech meter, I have to say that the build quality of the "fluke 17" extech is far better than the usual extech, and that the usual extech build quality is on par with or possibly below mastech.
Granted, the extech meter i have opened up was a dud. But it was SCARY. Connectors appeared to have been hand cut from legnths of socket strip. through-hole resistors not seated all the way. Halfassed reworks . . . .http://rubix.areb.org/gallery/20050917/img_0461_ahttp://rubix.areb.org/albums/20050917/img_0463.jpghttp://rubix.areb.org/gallery/20050917/img_0461?full=1
Obviously this was a defective unit that was reworked by someone. Badly.
But it's the little things that stick out, like the quality of the probe sockets. The high-amperage sockets are a lower quality socket. You'd hope they would be at least as high quality as the regular probe sockets, because more is demanded from them when they are used. But since they're used less frequently, they are lower quality.
This meter does not work. It also has a broken screen. I'm pretty much out of luck as the ebay seller i got it from wouldn't budge and paypal was no help. The seller disputes that the screen was not cracked when they packed it, but the evidence as i see it is that the screen was smacked HARD against something, and that couldn't have happened through 6" of packing material, so there's no way it's shipping damage. Paypal offered to refund if i shipped it back "in it's original condition" but i think it would be very easy for the seller to argue that the cracked screen was not the original condition so, I'm just screwed.
If you search ebay, the only Fluke 17b for sale is in hong kong, and they claim it is made by "Fluke China" -- but Fluke meters are made in Washington. I think they are either selling old stock that Fluke refused to buy, or they are using the Fluke name outside of their license agreement.
The Craftsman model number is 81437 (sometimes expressed as 34-81437 or 3481437) and in fact i did get one on ebay for $25, but i haven't unpacked it yet.
Fluke and Wavetek/Meterman meters are certainly better than mastech and extech. I won a Wavetek LCR55 for $27 on ebay and it is a very sensitive and stable piece of instrumentation. I have a "high end" mastech as well, and it works fairly well for most jobs but it's worthless for small AC measurements, making it's applicability for a lot of headphone jobs rather sketchy.
While setting up my M3 on sunday, my mastech was telling me that it was seeing 10mv of AC between IG and each output, with no input or output devices connected and the volume pot turned all the way down.
I didn't believe this at all, so i touched the probes together to see if it would zero-out, and it did not. It was still telling me i had 10mv out there. So i hit the 'relative' button to zero out that measurement, and went back to the M3
And it told me i had -10mv of AC on each output.
It has no trouble giving me accurate measurements of larger signals, but i think most people with cheap DMMs will have the same problem setting up an M3, which will be distressing for them since the setup instructions say that you have a problem if the probe reads more than .1mv in that state.