Musical Fidelity B1

[nairod156]

Oh crikey, that's me out of my depth. Think I'll have to give up on this one, but thanks all the same everybody for your help.

Ian

[Nakuser]

Some basic questions about the setup:
I do not have a distortion analyzer, but a scope and two voltmeters.
So setting it up would only be possible for me, to get the Kat´s voltage measures about the 0.47ohm resistors and trying to get the least DC offset possible.
Kat, you measured 23mA through the upper half and 30mA through the lower half? I do not have the schematic at hand, it is R63+R64 = 23mA and R65 + R66 = 30mA, correct?

I would like to build a larger transformer into the B1. Could anybody confirm, the original transformer is 2X25V / 80VA?
I would like to put paired 2N3055 into this amp. Somebody from the DIY-Audioforum stated, this is not needed for this design and mating was never performed by the MF Service. Correct?

All the best,
Nakuser

[redhunter]

Hi guys, i am new to this forum but i have been trying to get a circuit diagram for a B1 for years. So thanks for that.
I have had 3 of these and still have 2 currently. Many years ago i spoke to one of the engineers at MF and he told me to set the bias to 10 millivolts across each of the 4 middle 0.47R resistors and to leave it at that. You are right about them being somewhat unstable but it can be done with patience. I set mine to about 15 mv to give some room for drift. At this setting the amp runs fairly cool and still sounds great. I have 25 turn trimmers on mine which makes it easier.
The first one i had was an early production unit and had 4 little daughter boards around the bias set area. I had it a week and it sounded great but it used to get so hot that it smelled of burning paint! This was replaced with the one i have now and have had for 20 years. It has been heavily modified and has twin transformers. One for the drive and phono stage and the the original one for the power amp. It also has many upgraded parts and some circuit changes to the drive op amp.
Interestingly mine was never fitted with a Zobel network on the output either.
The phono stage on these is really good but mine has been disabled because i use an external battery powered one. This gives me another input to play with.
I have compared this with many much more expensive amps and have yet to find one that i prefer, which is strange really but MF must have done something right when they made these.

[Nakuser]

Hi Redhunter!
Do you remember the secondary voltages of the transformer?
I do not want to open my B1 again to measure them, but replace the transformer to a bigger one in the near future.
I also never found an amp, that could cope with the warm sound of the B1.
An A2 I tried recently, sounded horrible, but I guess, the B1 damps the treble a bit, which gives it a pleasant sound.
But: when adjusting the quimiscient current, I never get eqal values across the resistors.
My values, from the upper left to the lower right are:
30mV, 21mV (upper)
5mV 12mV (lower)
Of couse, those voltages are drifting
I always adjusted while trying to keep the DC offset at a minimum,
so I guess, it is impossble to get the same values for each resistor.
(R63 to R66) I also use precision trimmers.
Watching a 1kHz sinewave on a scope and raising currents equally for each side, until the "crossover dent" vanishes results in a permanent offset of 20mV.
All B1 I know popped when being switched on, except mine with 0vdc, so maybe 20mvDC offset is part of the design?

Even though I put in an Alps potentiometer, I still have differnt amplification
for the channels. Close before clipping level and the Alps at 12 o´clock, amplification is eqal. I guess this error comes from the Alps.
So I replaced R30 after the opamp with precision trimmers as well and matched signal level 9 o´clock, my preferable listening level. Hope, this was the right way to go...
What was your mods with the opamp? Would like to try them myself.
I also replaced the original opamp with a opa2134, but would like to know
the name of the original opamp, I simply lost the original one. Looking for the name given in the schematic on the web gives no results.
All the best,
Nakuser

[redhunter]

Hi Nakuser, in answer to your questions, the secondary of the transformer is about 2 x 25 volts AC. The original opamps in mine were TL082 but i now use OPA2134. I have tried quite a few over the years though. My main amp has about 20 mv offset in one channel and about 1 mv in the other. I dont worry too much about this though as it is quite difficult to get low offset and correct bias. I am happy if it is under 50 mv.
Both of my amps have poor channel matching and at one time i was going to put a decent pot in but decided not to as the original one is quiet. What i did was to lower the gain of the opamp to get less series resistance through the pot. At the same time i included a trimmer on one channel to match them at the volume i tend to use. This works very well and gives good tracking even at low volumes. Much better than the old method which was quite poor in this regard. I use several homemade phono stages and listen mostly to vinyl but these have about the same output level as CD. Consequently my volume tends to sit at between 10 and 11 oclock for normal listening and i rarely have to adjust it when switching between sources. I have considered bypassing the opamp and driving the power amp directly via a different pot value and with the level from CD or my phono amps could probably do it.
I have also changed the coupling caps to very rare high value polystyrene ones and to be honest the thing sounds so good that i doubt if i could improve it much further.
The power amp PSU now has a 25 amp bridge followed by 6800uf followed by the 0.47R resistors followed by 2 x 10000uf. all of the psu caps are bypassed with multilayer ceramics and the opamps have their own psu which is regulated and has around 30000uf of its own.
So you can see that it probably cannot go much further considering the case size!

[Nakuser]

Hi Redhunter!
Funny, the same mods (More caoacitance, bypassed phonostage, precision trimmers, same opamps) The Alps Blue Velvet pot I put in did not help at all with poor channel matching, so how did xou lower the gain?
I this a better method than replacing R30 (Voltage divider with R28) with a trimpot?
The transformers I can get for both amp and preamp are 2x30V/120VA or even 2x30V/160VA (But I think, those will not fit physically).
Too much voltage. (2X24V same VAsare the alternatives)
All the best,
Nakuser

[redhunter]

Hi Nakuser, i reduced the value of R28 on the diagram and left the divider on the output as it is. I was going to bypass that as well but i read somewhere that dividers like that can help to improve separation by reducing crosstalk. I still dont like it though. I chose to reduce gain because i wanted to reduce the influence of that pot in series with the input signal and because of my fairly high output sources. I have also rewired the output directly to replacement binding posts, with speaker cable and removed the headphone socket.
I had to do all this without a circuit diagram because MF would not supply one. Still, having got one now there is nothing else i wish to do. It is quite an unusual circuit and the power amp is probably best left alone.
One other thing, i have had to replace some output transistors on both amps to get decent bias matching as there was always one pair that did not adjust as well as the others.

[Nakuser]

Hello redhunter
thanks for all, I am in hurry so then quick question:
2x32V are too much?
All the best,
Nakuser

[redhunter]

It would be too much for my B1. But it would depend on the voltage rating of your smoothing caps. They would have to be of at very least 50 volts rating.
The 2N3055s and other transistors should be OK but it would be a bit of a gamble.
Are you trying to get more power from it, or just improve the dynamic response? It is already a very powerful and big sounding amp, way beyond what its actual power rating would suggest.

[Nakuser]

SHi Redhunter!
Silly me, I meant 2x30V at120VA or 160VA.
24V, same VA´s are the alternatives.
I replaced the smoothing caps, but have to take a look, wether it was 50V or 35 V Types. (2 10000uF I added are for sure 50V, but the four others, I do not remember)
I will have a look this evening. I would like to get more power.
My B1 clips with a 1kHz sine at full scale at already 1 0´clock, I would expect
more...

[redhunter]

I once tried using both of my amps in a dual mono set up. I was considering using just one pot to adjust the volume on them. This showed me that there was no advantage to using 2 power supplies and halving the work that each amp does and in fact i preferred the sound with just one amp. It sounded more coherent with better timing, which is one of the amps strong points. Good luck if you raise the PSU voltage. Personally, i would like to see 2 transformers and 2 bridge rectifiers in there, but the only way i could do that is by buiding an external PSU and i already have 2 many boxes!

[Nakuser]

Hi Redhunter!
I have 50V and 63V smoothing caps.So 120VA/2x30V toroidal transformer could be a replacement, but you think, there is no benefit in doing so?
One question about matching the 2N3055 output tansistors: Wil measuring
the hfe be sufficient?
All the best,
Nakuser

[redhunter]

Hi Nakuser, it has been a while since i played with this amp and i had forgotten that the transformer only has one secondary, not 2 x 25 volts but about 1 x 50 volts. As i said earlier if i did any more to the power supply it would be to put twin transformers or one with dual windings and use 2 bridge rectifiers instead of the one that is used.
You may get some benefit from raising the rail voltage but i didnt think it was worth it. If your amp is clipping at around 1 oclock just check that it is not the opamp that is clipping first as there is not much headroom here. My output stage does clip first but only just. If you are using the standard gain settings for the op amp you might find that clips first. Obviously if you reduce the gain here you will have to have the pot setting further round as mine is.
I cant remember clearly but i think i made up a little test jig that roughly simulated the working conditions, ie 10 mv across 0.47R to select 2N3055's by HFE. They did vary quite a bit. This solved the problem of poorly matched bias settings to large degree though. I think they just selected transistors by batch when new, as the engineer that i spoke to at MF implied that they replaced them all when they had blow ups! The transistors parameters also seem to change with age and heat but as i said mine runs fairly consistantly cool now and has for years. By the way i checked bias a few days ago and it needed no adjustment so that was a relief.

[Nakuser]

Hi redhunter!
Thanks a lot for all the explanations. I have to open my B1 again, because I think, it was 2X25V though a center tap. This makes more sense.
All the best,
Nakuser

[redhunter]

Sorry, Nakusa i meant 50 volts with a centre tap as opposed to 2 x 25 volts which i would have preferred. Have you heard any of the other older MF amps by the way?

[Nakuser]

Hi Redhunter,
well, iI tried a Vincent SV-121, SV-231, Shanling SLM-A40, the latter Class A, and a MF-A2.
The SV-231 was the best among them to my ears, but needs too much power.
The other ones were too bright, the SV-121 digita volume control clipped with the 3V input from my DAC, which delvers studio level.
Maybe, the brightness and lack off bass is more close to the original signal,
but I prefer the B1 with it´s soft sound. I already use it for 20 years.
I hesitate to try the MJ15003 as output transistors, which should be more linear tha the 2N3055.
Do you remember the value of R28 for lowering the gain of opa2134?
All the best,
Nakuser

[redhunter]

Hi Nakusa, i have put a 16k across R28 on one channel and a 20k trimmer across the other R28. This has enabled me to match the output of each channel to the volume pot at the setting that i most use. It does give a nicely linear response almost down to the fully off position as well as allowing me to use less of the pots resistance. I think what you have done is fine but i always felt that the thing had a bit too much gain. I have a tuner which was very loud at the 9 oclock position and both of my CD players were almost as loud. The onboard phono stage needed to be at around 10 to 11 to give similar volumes. Now the pot sits at around 11 oclock most of the time. Of course it also depends on the efficiency of your speakers.
One other thing that i have done which mainly benefitted the phono input, was to use a 3 core mains lead and earth the amp. I also have an inline mains filter fitted to the lead outside of the amp. If you would like any more info, dont hesitate to ask or PM me.

[Nakuser]

Hi Redhunter!
Great, I will give the 26K mod a try. maybe, with the alps, a 20ktrimmer is not needed.
I have the same experience, tuner is very loud, phono stage was loud (I did cut the traces and use the input now for my dac, I have no record player anymore) But Voltages/Loudness from Line-sources can vary very much.
My Sony Player delivers 2,5V at fulls scale, an old nakamichi 2.1.V the DAC even 3V.
I never tried modifying the mains lead: I live in an old house, where safety ground an mains ground are connected, so there is no "quiet" safety ground to absorb currents.
Thanks for your offer, I will definately come back on this...

Oh, btw, what foil capacitors were attached between speaker signal and ground at the output jacks? They are not in the schematic, removed mine years ago and forgot the values. But I guess, they are not needed anyway, because maybe they were used to prevent oscillation, but "newer" 2N3055 have a transit frequency of about 2mHz, as far as I know...
All the best,
Nakuser

[redhunter]

My oldest amp never had any caps fitted to the output, and i had it from new. That is the one that i have been using for the last 20 years. I have another that was originally purchased for a second system while it was cheap. It too sounds good but has had far fewer mods than my main one. It had a couple of mylar caps on the output with series resistors but like you i removed them. I think that they were about 100 nanofarads but i cant be sure.
I would not describe the sound as "soft" though. Mine is hard hitting and gives a huge soundstage that fills the room from walls to ceiling. It projects instruments away from the speakers and has a seemingly unlimited dynamic range. That is how it sounds to me with my speakers, which are Epos ES11 and my LP12. CD is more confined to the boundaries of the speakers and does not project as much. B1's do sound dark though but i like that and the treble is not too fierce. Bass is full, rich and fruity especially with my LP12.

[Nakuser]

Hi redhunter,
well, sounds like we will keep our beasts for the next 20 years.
I use 3A Audio Design 2-way speakers, which I purchased with the B1 twenty years ago.
The company is now called Reference3A, they moved from France to Canada and changed the name, but the design has not much changed in 20 years.
I will never give them up as well. Two years ago, the only source for the tweeters was a shop in the Netherlands, I guess, I got the last unused ones...
Same with other equepment, for my Nak player I bought two KSS-123A laser units as spares while they were cheap and available, same with the Sony CDP.
Especially for players, I never thought they could become vulnerable vintages,
but as sound quality might have improved, built quality has declined...
All the best,
Sal