Some years ago i repaired a BM1000. There are lots of Roederstein capacitors inside, red plastic caps. They were 40-50 years ago pretty good, this time is gone. Most of them are either dead or they spread acid on the board, they are cracked.  Replace them all with Panasonic FC and FR types. Surprisingly a lot of the soundpath is done with Tantalium Caps. This is probably one of the worst sounds ever. Tantal in the soundpath is no pleasure, Tantalium reacts different on the frequency, the inner resistor is not linear, it's curved. Replace them with bipolar MKS WIMA Foil caps. MKS-2, MKP... whatever fits. After this mod the BM1000 plays in a total different sound level. Much better. The manual switches are a pain with hard grease. Clean it with contact cleaner and mechanical Oil like WD40. Some of the switches got stuck in the old grease. After cleaning and oiling all switches became smoth and reliable. Good Luck!  

That is GREAT info, thank you! I won't even power it up before working on it: I found reference that one of the caps in the power supply tends to explode. So I'll do all that before plugging it in. Guten tag!

What, in your opinion, would be a better RIAA curve? Martin

I've seen restorations where new & better tolerance parts have been used to make the RIAA circuit more accurate, as well as improved the phono preamp.   So I was wondering if anyone had done that with the 1000

Login to view image Login to view image The above plots are of the "deviation from ideal" for the RIAA amps of two different Beomasters. Areas of "improvement" could be to better match the curves to the ideal curve, make the left and right channels match better, or try to match the curve to a particular cartridge. Regardless of what your goals are, one will need a way of measuring the circuits to know if you achieved them. Glitch  

Hey those are interesting, thanks! I can't figure them out though: are the 2 colors the left & right channels?   Or is one of the colors supposed to be the RIAA baseline? Because it doesn't appear to match between the 2 images, although the graph scales are different. Apologies if I'm missing something obvious

The green/red colors are for the left/right channels. The test is of a series of (nearly perfect) RIAA weighted sine waves being fed into the circuit board and the output measured. Each vertical line is the result at a given frequency. All of the bars at the same height would indicate that the circuit perfectly implements the RIAA curve. The deviation from a flat response is the "tonal coloration" of the circuit. Differences between the green and red curves indicate that each channel has a different coloration. This kind of plot can really highlight differences between circuits or issues within a given board. For example, the second plot was for a board that was recently recapped using fresh, high quality Panasonic electrolytic capacitors. This shows that you can't just blindly swap parts and expect perfection. There may be variation in the new parts or another underlying issue in the other components. Another example would be if someone suggested component values based on the first circuit and you were working on the second circuit. I wouldn't even want to try to guess what the final result would be. Glitch    

Wow!! That's quite amazing! I wonder if testing it right after the RIAA & before the rest of the circuitry would be the best way to assess? Otherwise tone controls, noise & interference from other things in the circuit would affect results? I guess one could do both, to figure out what's going on

The curves above are for just the RIAA stage. You are correct in thinking that adding more stages would complicate any analysis. Testing smaller subsets also allows one to tailor the test to the component being tested. For example, testing the power amp stage is a quite different test than testing the RIAA or pre-amp stage. I prefer to be data driven when doing repairs and modifications. I don't think that there is anything particularly wrong with the swap parts and listen methodology if one likes the results. I just think that I reach my goals more quickly and with more confidence with test data. Glitch