Musical Fidelity B1

[Kat Manton]

Hi,

I've recently been given one of these hi-fi amps in a state of deadness.

I think I've fixed it (failed output transistors) but I'm struggling a bit with the output stage bias adjustment. It's a class AB amplifier.

Does anyone happen to know the procedure for adjusting the bias?

I haven't got very far with inspired guesswork and twiddling, apart from being able to obtain 0V DC offset on the speaker terminals, not have it threaten to go into thermal runaway... and produce audible and visible distortion on a 1kHz sine wave when the volume is advanced to "reasonable"...

Both channels are behaving similarly.

The circuit (attached) is interesting, to say the least.

Cheers, Kat

[Steve_P]

Check R63-66. Only half an ohm but it sometimes does matter.

Set those pots between C24 and C25.

Remember it's Fidelity....

Cheers,

Steve P.

[Kat Manton]

Hi,

Just to clarify, this is a stereo integrated amplifier made in the early nineties by Musical Fidelity, a British hi-fi company founded in 1982. There's no connection that I know of between this company and the consumer electronics company which shares half its name.

There's some information on the company's archive page for the B1.

Searching for information about this amplifier turns up a few results, including forum posts by people who've already contacted Musical Fidelity seeking this information to no avail. I may try contacting them myself, just on the offchance.

The two presets marked 'set bias' on the diagram do, indeed, adjust the bias. What I need to know is what to measure, where to measure it and what measurement to aim for when adjusting.

So far, I've obtained the best results by feeding a 1kHz sine wave into the unit, listening to the output and adjusting each preset alternately for minimum audible distortion, with a final tweak of one preset to reduce the DC offset to zero. At each extreme of each preset, distortion is obvious, reducing over a range in the middle. Both adjustments are interdependent.

However, I'm aware that I can't hear distortion below a certain level. I can't measure it either, as my Sound Technology ST1700B's oscillator isn't currently working in 'low distortion' mode.

Alternately adjusting each preset to the middle of the range which produces least audible distortion also leaves the output stage running quite hot. This may be correct; Musical Fidelity's class A amplifiers had a reputation for running too hot to touch. It's possible that this class AB design is intentionally biased a little more towards class A than many.

After adjusting one channel by ear, I've taken voltage measurements across the 0R47 load resistors R63 to R66 and calculated the current through them. The voltages are approximate and drift as the transistors warm up.

80mV across R63 - current 170mA
50mV across R64 - current 106mA
10mV across R65 - current 21mA
30mV across R66 - current 64mA

One observation is that these measurements are all over the place; I'm beginning to wonder if Musical Fidelity used matched transistors. Due to lack of transistors I only replaced the dead ones; which isn't what I usually do. Replacing all four in each channel with ones from the same batch is one of the next things I'll try.

I'm very inclined to match pairs up, then use each pair in parallel.

It's also possible that the 0R47 (5%) resistors have drifted in value. A few have discoloured a bit though they all measure ok.

I should soon have more than enough 2N3055s to replace all eight and replacements for the resistors though.

In the absence of any definitive procedure, my intuition suggests adjusting by ear while aiming for somewhere around 75mA. The two transistors passing over 100mA are getting hotter than I'd like. The other two are warm, closer to the temperature I'd expect under quiescent conditions.

Unless anyone has any better ideas, that is

(I'll also try, yet again, to get hold of a service manual for my ST1700B.)

Cheers, Kat

[mhennessy]

Hello Kat,

I know quite a bit about the A1 (see my website), but until now, I've never seen the schematics for the B1. Thank you for posting them.

I do know that the B1 was, like the A1, a Tim de Paravicini design. He, as you may know, is a very innovative engineer who rarely follows convention. Hence, this amplifier might seem somewhat "quirky" at a first look. But my dealings with the A1 have taught me to look beyond this...

The symmetry is also present in the A1 design. You could remove each half of the amplifier (replacing it with a current source or even a low value resistor) and it would work! Understanding that this is, in effect, two separate amplifiers will help to work out how to adjust those preset resistors.

Looking at the top half, the LTP uses CFPs (complimentary follower pairs), but this doesn't change the basic operation. The output is buffered by EF T16, then T18 is the VAS. In a "normal" amplifier, you'd find a buffer between this and the output devices, but not here. Actually, T18 combines with T20/T22 to form a CFP, hence D1, which is a Baxendell diode.

The bottom half is the same, the only difference being T17, which is an emitter follower. Thus, the bottom half is Darlington rather than CPF. Presumably this was done to save cost...?

Anyway, key to the adjustments is the feedback resistors R50 and R51. Note that NFB is taken from before the 0.47 ohm resistors - not from the junction. Without the preset resistors, there would be 0V at the base of T10 (and T13) because the input is held at 0V by R33, and therefore, no standing current in the output stage (and crossover distortion). So an offset is introduced so that NFB can maintain a small voltage across the 0.47 ohm resistors. In theory, this NFB action should stop the unit entering thermal runaway, but I offer no warranty

As you've found, it's important that the offsets match each other so that the net DC offset is close to 0V. So I wonder if one should measure the voltages across R64 and R63 with respect to ground? This would make sense were it not for the extra transistors, and note that these are not included in the feedback! Yet their contribution to the standing current should have an effect on the THD, so I would guess that on the production line, these would have been lined up using a distortion analyser as well as a voltmeter.

Looking at your numbers, they don't quite add up. About 6mA might come from R58, but 15mA is "missing" somewhere. Did you have a load connected when you made these? Perhaps these are just the errors caused by the tolerance of the 0.47 resistors...

I don't know if they would have gone to the trouble of matching the output transistors for this low-cost model. I've been told that they didn't do any selection for the A1, but they did re-label the devices to make them appear more "mysterious"!

Sorry this has been a long post, and sorry for not giving you an authoritative answer. But hopefully it's helped explain the circuit - as I say, it shares much in common with the A1, and there were quite a few things I missed until T de P got in contact and explained them to me. It might be worth contacting him (lookup "EAR", his company), but I'd understand if he didn't want to respond after 25 years. Another bet might be Jez Arkless (email on my site), but he's running a business and understandably doesn't have much time to give free advice.

If you do get an answer, do post it here - I'd love to know!

Best regards,

Mark

[Kat Manton]

Hi Mark,

Thanks very much for the explanation. I'd found the circuit for the A1 (on your site) and noted the similarity of the two designs.

Incidentally, as I found the B1 schematics on the 'net (I can't remember where, but it took a lot of searching) and they were freely available, I don't see any reason why you shouldn't add them to your site if you like. At least they'd be easier to find for anyone else attempting to service a B1.

I've now come up with a way of adjusting it. The biggest difficulty was attempting to watch two measurements while adjusting the two interdependent 'set bias' presets.

This is much easier to perform using a 'scope in X/Y mode, X connected to the top end of R64 and Y to the output.

I start with both inputs grounded and centre the dot. Then I shift it left (negative) by the desired voltage on R64. If I now adjust both presets to bring the dot back to the centre, both measurements will be correct.

It's still tricky and the dot tends to drift, but it's the easiest method I've come up with so far.

I'm still waiting for the postman to deliver some bits; at the moment, one channel's fitted with two 'Pecor' 2N3055s in the lower half, two marked 'A1000N' in the upper half and somewhat fried-looking 0R47 resistors. (The other channel's missing all four transistors now...)

I've also got around the problem of the dodgy oscillator in my ST1700B with the aid of a computer and high-quality sound card. A 24-bit 1000Hz sine wave generated with a sample rate of 96kHz measures 0.004% when the card's connected directly to the analyser; I think that'll suffice.

Using the 'scope, I can fairly quickly set the voltage at the top end of R64 to anything I like while keeping the DC offset at zero (with no signal), then measure distortion.

So I think it'll be possible to figure out a reasonable bias setting without it taking forever

A few distortion measurements made with different bias settings suggests ~25mV (~50mA through R64) may be "about right" but I'd like to fit new parts (and have both both channels working) before reporting any further findings.

Cheers, Kat

[dseymo1]

Brilliant bit of lateral thinking to use a scope like that!
I must admit that I rarely consider scopes for pure DC measurement.

[Kat Manton]

Hi,

Originally Posted by mhennessy

Looking at your numbers, they don't quite add up. About 6mA might come from R58, but 15mA is "missing" somewhere. Did you have a load connected when you made these? Perhaps these are just the errors caused by the tolerance of the 0.47 resistors...

I think the errors may just be down to the time between each measurement. Both the DC offset and voltage on the top end of R64 drift about, never quite settling down.

Originally Posted by mhennessy

I don't know if they would have gone to the trouble of matching the output transistors for this low-cost model.

Probably not and neither did I.

Apart from them being unmatched, each device has its own heatsink and they all end up operating at different temperatures. I'm wondering if this is connected somehow to the drift I'm seeing.

Anyway, cutting a long story short, I've now replaced all eight output transistors and experimented with different quiescent currents.

This is how I have it operating at present, these are the voltages in mV across the 0R47 resistors, 'upper' and 'lower' as per the schematic, measured with no load connected.

It'd been left on for at least 30 minutes and had vaguely stabilised, not that it ever really does...

L: upper: 21, 24; lower: 41, 4
R: upper: 27, 13; lower: 36, 7

About 5 minutes later...

L: upper: 20, 23; lower: 40, 4
R: upper: 25, 13; lower: 35, 8

This works out as roughly 80-100mA total current for each channel, shared rather unevenly.

Distortion for both channels measured at 1kHz is 0.03% at 1W and 0.12% at 3W and 10W into 10R (the closest to 8R I could find.)

The specification (attached) states "less than 0.5%". That seems high for a solid-state amplifier but specifications never tell the whole story. Music is a bit more complicated than a 1kHz sine wave...

It's now connected to Celestion Ditton 44s and... well, I don't want to end up sounding like a magazine hi-fi review but 'detailed', 'open', 'transparent', 'clean' and 'tight bass' might be mentioned if I let myself get carried away.

"It works", "it runs a bit warm but hasn't caught fire yet" and "I'm thoroughly enjoying listening to a wide variety of music" will do

Cheers, Kat

[mhennessy]

Hi Kat,

Good news - glad it's working again. I'm not surprised at the THD results - none of MF's amps from this era were particularly clean, but the unusual circuit layouts resulted in a particular spread of harmonics that gave a "nice" sound - which I suppose is important if you want to stand out from the crowd. Certainly, my A1 was highly enjoyable with efficient speakers...

Spotted in your earlier post that some devices were labelled "A1000N" - that's the re-labelling that I mentioned before. For the A1, the 2N3055s became "A1N" and the MJ2955s were "A1P". IIRC, the A1000Ns ended up in the A100 amp, but again, they were just 2N3055's.

And I also appreciated your X-Y setup - very neat!

Best regards,

Mark

[Kat Manton]

Hi Mark,

I've been experimenting some more, I just couldn't leave it alone

I wasn't too happy with it running hot due to those rather high standing currents. It's supposed to be a class AB amp but I'd got it biased moderately into class A.

So I looked for another way to set it up.

Originally Posted by mhennessy

As you've found, it's important that the offsets match each other so that the net DC offset is close to 0V. So I wonder if one should measure the voltages across R64 and R63 with respect to ground? This would make sense were it not for the extra transistors, and note that these are not included in the feedback! Yet their contribution to the standing current should have an effect on the THD, so I would guess that on the production line, these would have been lined up using a distortion analyser as well as a voltmeter.

That's what got me thinking. I'm now fairly sure that it must've been originally set up with a distortion analyser and can't really be done without one.

This is the latest setup procedure I've devised.

Start with all 'set bias' presets set for minimum bias.

With the front towards you, from left to right:

1. Fully clockwise
2. Fully counter-clockwise
3. Fully clockwise
4. Fully clockwise

(Yup, one of 'em works the opposite way to the other three, just to add to the fun...)

For one channel:

Feed in 1kHz sine wave, adjust volume for 1V across 10R load (100mW.)

Alternating between the two presets, slowly increase the bias while keeping the DC offset somewhere around 0V.

The volume control will need slight adjustment to maintain 1V output.

Watching the output on one trace and the distortion output of the analyser on the other trace, slowly but surely the crossover distortion can be seen to fall.

Eventually, it's possible to minimise the crossover distortion while still having some; in other words the amp isn't biased into class A.

There's a slight dip in distortion on the meter at this point. If the bias is increased beyond this point, the distortion rises slightly then falls off rapidly as the amp is now running in class A.

With a bit of careful tweaking it's possible to get this 'dip' to coincide with 0V DC offset.

Repeat for the other channel.

After setting it up using this method, I've measured the voltages across the four 0R47 resistors.

I now have what seems like a more reasonable 23mA total through the upper half and 30mA through the lower half, on both channels. The discrepancy will be due to R58.

Distortion is still within specification (actually 0.12% up to around 15W), it's running a lot cooler and doesn't sound any different.

I'm now of the opinion that it isn't possible to set up a B1 properly without a distortion analyser. It must be how they were originally set up; just tweaking the bias presets for some voltage measured somewhere doesn't really work.

This is unlike the more 'typical' class AB amplifiers I've worked on. You usually get two presets, one for DC offset and the other for bias. Setting up involves nothing more than adjusting the bias preset for some mV across a resistor and the DC offset preset for zero.

I think that's it... Let's see how long I can leave it alone for now...

Cheers, Kat

[mhennessy]

Hi Kat,

Thanks for the detailed post. I reckon you've got it spot-on using that technique - it's probably better than new now

I'm wondering - will the thread always be available? What happens when it gets archived? It would be a shame to lose this information - would I be allowed to copy it and add it to my website (with appropriate credit, of course)?

Best regards,

Mark

[Nakuser]

Wow!
Now this is what I was looking for!
Got my B1 somehow biased with the aid of an italian forum last spring, but I see now there is still something to do.
I just found a german company which offers paired 2N3055, so I might buy some and try to adjust my B1 in March.
Some years ago, I had to replace the transistors 15-18 with newer types, after the amp blew.
I cannot read the handwriting so well, what are their names?
I would like to buy the originals again, should they also be selected/paired?
Last year, I tried to find a replacemet for the B1, but was always disappointed. The worst experience to my ears was a MF-A2, which sounded too sharp compared to the B1...
So, got this amp for almost 20 years and I guess, it will run the next 20 years.
One backdraw seems that my B1 clips too early for 32W (around 10 o´clock
on the volume knob.)
BTW, the AC Voltage from the transformer is about 25V?
All the best,
Salar

[nairod156]

Hi Everyone,

I found this thread when searching online for a solution to my B1 amp problem and wonder if you might be kind enough to give me some advice.

A while ago my beloved B1 stopped working. I have extremely limited knowledge of electronics and although I managed to identify a burnt out transformer I am at a complete loss as to how to work out what rating transformer to replace it with.

I did approach a transformer company but was told they would have to reverse engineer the old one and the bill would be around £150! Can't afford that, so any help anyone could give to point me in the right direction would be much appreciated.

Many thanks,

Ian

[Nakuser]

Hi nairod,
reverse engeneering is not needed. I am also not a technician, but as far as I know, the transformer has simply 2x25V (maybe 2x27V) secondaries.
Looking at the size, I guess it is about 80VA, but the others here can give
more exact answers... So wait a bit.
But finding a replacement transformer should be easy, these are relatively common values.
All the best,
Nak

[Nickthedentist]

But, you also have to consider why the old transformer failed

Nick.

[Growlerman]

I remember a story my local (at the time) hi fi dealer told me regarding the B1 amplifier. He was a MF stocklist and was offered the new B1 amplifier to demo to his customers before the B1 was released. The B1 MF supplied sounded absolutely brilliant that a lot of orders were placed with MF. The dealer was not allowed to sell the demo B1 as it had to go back to MF. When the production B1's came through they sounded very different to the demo B1.

[Nakuser]

What about the schematic circling through the web?
Maybe it is from this demo?
Or do you think, they just changed the less critical parts to cheaper ones?

Well, MF is not very user-friendly giving information about discontinued models.

[Michael Maurice]

There is no protection on this amplifier apart from the mains fuse, this should blow if there is a fault such as s/c transistors, but if someone put the wrong fuse in, it could cause the transformer to burn out.

I understand it was common for some 'exotic' circuit designers to not put in any protection to keep the signal path 'pure'.

[nairod156]

Hi All,
Thanks for the response. I've just had a look at the fuse and it's got what looks like T1A stamped into it. Does that sound like the correct fuse for the amp?

Very interesting story about the MF demo model Growlerman. What a crafty dodge if it's true, and I can't think of any other reason they would insist on it being returned to them.

And thanks for the technical details Nakuser. Gives me hope that I may get my amp going again after all!

[nairod156]

So do you think it would be foolish to replace the transformer without knowing what made it fail in the first place?

[Michael Maurice]

Absolutely!

At the very least, you should remove the circuit board from the case and check the output transistors for short circuits. Each channel uses four 2N3055's so if one has gone, all four must be replaced. Also check the bridge rectifier to see if that has gone short circuit.

Any of these could have cause the transformer to fail.