Jump to content

Bose 901 Grill Removal Details


tysontom

Recommended Posts

Help... Bose experts:

Question 1: How are the grill cloth and frames on a standard-version Bose 901 (Series I and II) assembled? How are they attached such that they can be removed carefully and replaced without damage? I have a beautiful pair albiet with one renegade burned-out driver, and I need to get inside to make the repair! I have the replacement driver, but I have not removed the grills for fear of damage until I know how Bose put them together. What comes off first, etc. Are there staples? Glue?

Question 2: Who is expertise in rebuilding (i.e., replacing capacitors, etc.) the 901 Equalizer boxes? Can the early equalizers be returned to Bose for service? All my three 901 (Is and IIs) units work, but I'm sure the capacitors have begun to leak by now.

Thanks for your help! (Soundminded, I know you are an expert, but I think you have a pair of the Walnut-faced versions, but this pair in question is the original-style). I'm sure you have good data on the disassembly procedure for these models.

--Tom Tyson

Link to comment
Share on other sites

Hi Tom

They are held in by staples. The staples are white. You have to look very carefully because they are nearly invisible against the weave of the grilllcloths. Pry them out gently with the tip of a knife, a screwdriver and needlenose pliers and the masonite with the grill cloth should come off easily. If the front and sides have been replaced with walnut panels, those are held in with double sided tape. They can also be removed to access the front driver but be very careful not to damage the wood as it is thin.

BTW, in series one and two the drivers came from CTS and were sorted by Bose into three batches, each enclosute had matching drivers from the same batch. In series one the sorting was done manually. In series two, it was done by computer. I've never taken any of my drivers out so I don't know what kind of markings are on them and I have no idea if there is any stuffing inside. Given about a 7 db peak at around 250 to 500 hz, I'm guessing there isn't any but I'd be interested to find out what you see when you open yours.

Instead of replacing the caulking which had dried out, I used clear GE silicone caulking to seal the edges between the frames of the drives and the wood. I also put a daub on each screw head. It worked very well, the most important thing is not to get any on the suspensions or cones.

Bose considers the series one and two non repairable. I have no reason to believe that the caps on mine are bad and have no intention of replacing them. they were all 1% and 5% parts. BTW, I had a hum problem. It turned out to be the result of using a press fit for the grounding of the input and output jacks, there were no soldered electrical connections to them. Once I soldered ground wires to all of them, the problem went away and I never heard it again.

Link to comment
Share on other sites

Hi Tom

They are held in by staples. The staples are white. You have to look very carefully because they are nearly invisible against the weave of the grilllcloths. Pry them out gently with the tip of a knife, a screwdriver and needlenose pliers and the masonite with the grill cloth should come off easily. If the front and sides have been replaced with walnut panels, those are held in with double sided tape. They can also be removed to access the front driver but be very careful not to damage the wood as it is thin.

BTW, in series one and two the drivers came from CTS and were sorted by Bose into three batches, each enclosute had matching drivers from the same batch. In series one the sorting was done manually. In series two, it was done by computer. I've never taken any of my drivers out so I don't know what kind of markings are on them and I have no idea if there is any stuffing inside. Given about a 7 db peak at around 250 to 500 hz, I'm guessing there isn't any but I'd be interested to find out what you see when you open yours.

Instead of replacing the caulking which had dried out, I used clear GE silicone caulking to seal the edges between the frames of the drives and the wood. I also put a daub on each screw head. It worked very well, the most important thing is not to get any on the suspensions or cones.

Bose considers the series one and two non repairable. I have no reason to believe that the caps on mine are bad and have no intention of replacing them. they were all 1% and 5% parts. BTW, I had a hum problem. It turned out to be the result of using a press fit for the grounding of the input and output jacks, there were no soldered electrical connections to them. Once I soldered ground wires to all of them, the problem went away and I never heard it again.

Soundminded,

Thanks very much for your detailed answer! This is exactly what I was looking for, and it will help me a great deal. I will let you know what I find inside; I do believe, however, that there is fiberglass inside the Series I and II versions in that those models (I'm not sure when the changeover to the ported enclosure occurred) are acoustic-suspension enclosures of sorts, and being stuffed with fiberglass would help to dampen the speaker as well as lower the system resonance, even though everything is equalized.

Actually, as you well know, the Bose 901 has been highly criticized for its sound, yet when a pair is properly mounted in a good room, and powered adequately, the end result (except for the highest frequencies as you noted) is very pleasant and life-like. There is an effortless and smooth quality to the sound of a full orchestra when reproduced by the early 901s that most conventional loudspeakers seem to lack. Whether people agree or disagree with the direct/reflecting sound design of the 901, Dr. Bose clearly knew what he was doing.

--Tom Tyson

Link to comment
Share on other sites

Soundminded,

Thanks very much for your detailed answer! This is exactly what I was looking for, and it will help me a great deal. I will let you know what I find inside; I do believe, however, that there is fiberglass inside the Series I and II versions in that those models (I'm not sure when the changeover to the ported enclosure occurred) are acoustic-suspension enclosures of sorts, and being stuffed with fiberglass would help to dampen the speaker as well as lower the system resonance, even though everything is equalized.

Actually, as you well know, the Bose 901 has been highly criticized for its sound, yet when a pair is properly mounted in a good room, and powered adequately, the end result (except for the highest frequencies as you noted) is very pleasant and life-like. There is an effortless and smooth quality to the sound of a full orchestra when reproduced by the early 901s that most conventional loudspeakers seem to lack. Whether people agree or disagree with the direct/reflecting sound design of the 901, Dr. Bose clearly knew what he was doing.

--Tom Tyson

Here are a few observations;

The inventor Dr. Bose used a small enclosure for his acoustic suspension design to push the resonant frequency up. He needed to get it above 180 hz where he said the phase shift associated with resonance becomes inaudible. He exploits the linear falloff of amplitude with frequency below resonance with a complimentary equalizer. BUT!...his equalizer has only a 6 db per octave rise while the acoustic suspension speaker has a 12 db per octave falloff. My experiments convince me my system has a resonance in the range of 250 hz of around 8 db. e/e magazine reported theirs at 500 hz and a 7 db rise. To achieve flat response, whatever the effective response in the room is, it requires further equalization. My calculations show that to achieve the same level at 30 hz you'd get at 1khz at one watt would requre the 30 hz signal to be about 600 to 1000 watts. This would be very loud in most rooms. The system can only hande a rated 270 watts per enclosure so 3 pairs in parallel driven by your Crown Reference II amplifier should do a great job. You should expect usable response down to the region of about 23 to 26 hz with 10% THD. This will put the three parallel pairs basswise in the same league as AR9.

The system lacks the capacity to reproduce the highest octave. The inertial mass of the drivers makes them very poor tweeters. Even if it could reproduce it, its single forward driver would produce high frequencies only directly on axis. This combined with its upper bass resonance is the reason why IMO audiophiles have not accepted it. Knowning this, it still took me three years of experimentation to achieve the redesign goals I set for it. The problem is not straighforward if you are a very critical listener and I have not and will not publish the mathematical model I use for all my analyses any time in the forseeable future. The re-engineered system comes as close to my idealized model for its function as I can probably get without spending a lot more money designing a new speaker from the ground up. Needless to say I am satisfied with the results.

Link to comment
Share on other sites

Here are a few observations;

The inventor Dr. Bose used a small enclosure for his acoustic suspension design to push the resonant frequency up. He needed to get it above 180 hz where he said the phase shift associated with resonance becomes inaudible. He exploits the linear falloff of amplitude with frequency below resonance with a complimentary equalizer. BUT!...his equalizer has only a 6 db per octave rise while the acoustic suspension speaker has a 12 db per octave falloff. My experiments convince me my system has a resonance in the range of 250 hz of around 8 db. e/e magazine reported theirs at 500 hz and a 7 db rise. To achieve flat response, whatever the effective response in the room is, it requires further equalization. My calculations show that to achieve the same level at 30 hz you'd get at 1khz at one watt would requre the 30 hz signal to be about 600 to 1000 watts. This would be very loud in most rooms. The system can only hande a rated 270 watts per enclosure so 3 pairs in parallel driven by your Crown Reference II amplifier should do a great job. You should expect usable response down to the region of about 23 to 26 hz with 10% THD. This will put the three parallel pairs basswise in the same league as AR9.

The system lacks the capacity to reproduce the highest octave. The inertial mass of the drivers makes them very poor tweeters. Even if it could reproduce it, its single forward driver would produce high frequencies only directly on axis. This combined with its upper bass resonance is the reason why IMO audiophiles have not accepted it. Knowning this, it still took me three years of experimentation to achieve the redesign goals I set for it. The problem is not straighforward if you are a very critical listener and I have not and will not publish the mathematical model I use for all my analyses any time in the forseeable future. The re-engineered system comes as close to my idealized model for its function as I can probably get without spending a lot more money designing a new speaker from the ground up. Needless to say I am satisfied with the results.

My amp is the Crown Reference I, which is the more powerful of the two, but the power rating of either amp is largely academic. These are powerful, conservatively rated amplifiers. As for the Bose 901 power rating, I suspect that the 270-watt rating is the continuous, steady-state power level at mid-frequency (probably 1kHz), not the peak-power rating. In other word, I think the 901s are capable of sustaining much greater levels of peak power without distress.

I have read that the harmonic distortion in the bottom octave is somewhat higher in the original 901 than comparable, high-quality acoustic-suspension loudspeakers of that period (such as the Advent and AR-3a). I believe that Julian Hirsch noted this with his tests, but I have never seen an actual comparison at more or less identical acoustical output levels. I have not heard what the distortion level would be in the new-style 901s. Although clean, low-distortion deep bass is a hallmark of AR speakers, I don’t think that small orders of distortion would be particularly objectionable unless one listened to strictly organ music on a regular basis. In this case, the higher distortion level for the 901 might be audible; however, the 901 is equalized flat down about a half-octave lower than the AR-3a, so the impression can be of more deep-bass output from the 901. For example, bass-drum reproduction is authoritative from the 901.

I am not so sure that it is an inertia issue with the highest frequencies. The little drivers are not ideal tweeters, for sure, and they begin to roll off pretty soon in the treble, but they could easily be equalized to respond up to 20kHz if Bose had wanted that, but I don’t think he wanted that particularly. Quite frankly, if the speaker has strong and smooth output up to 13-15 kHz, most people will find that to be realistic and fulfilling, and the lack of the top-most octave (not heard by the bulk of listeners anyway) is probably more of an academic issue than a true acoustical problem. Once again, I think that Bose carefully calculated what was heard in a concert hall, and he duplicated it fairly successfully in the home listening environment. The very high frequencies are attenuated (absorbed) in a concert hall anyway, so the 901's high-frequency roll-off might not be greatly felt. In the end, Bose succeeded in designing a speaker that has realism and smoothness in reproducing orchestral sound, and it was done pretty much at the expense of absolute flat on-axis response in an anechoic chamber or with perfect on- and off-axis response in the high frequencies. Somehow, the reproduction is more effortless than with most conventional speakers, probably because all of the nine tiny drivers are never really pushed hard, and because the sound is bounced around the room in a carefully calculated manner. I not partial to the sound of the 901, but I have always objected to the excessive criticism of it. The speaker is much better than many people think, and it has a rightful place being one of the great contributions to high fidelity.

As for my pair(s) of 901s, I intend to restore them, run some tests and then probably sell them.

(1) Information on the Crown Amps

(2) Showing the fiberglass inside the Bose 901 Series II

--Tom Tyson

Link to comment
Share on other sites

Interesting about the fiberglass. Are the drivers 72 ohms wired in parallel or 8 ohms with three in series and each series combination in parallel with the others to get back to 8 ohms?

The absorption of high frequencies in a concert hall is a dynamic event. The direct field has the full high frequency compliment that each musical intrument radiates in the direction of the listener. As the sound reverberates, the high frequency component is absorbed by both humidity in the air and the boundary materials of the hall selectively, that is high frequencies are absorbed faster than middle and low frequencies. The RT or reverberation time which is the time required for sound levels to fall by a factor of one million, that is 60 db in a typical modern concert hall is about 2 seconds at 1 khz, and about 1.2 seconds at 8 khz, the limits of the data I have. In older halls it is even longer, as much as 3 to 3.5 seconds at mid and low frequencies. This makes newer halls and renovated halls less reverberant increasing their clarity at the expense of reverberation (all other things being equal.) Because the spectral relationship of fundimentals and harmonics is a dynamic phenomenon which affects the perception of tonality, it is impossible for home sound systems using the current paradigm to reproduce the sound of musical instruments as they would be heard at a live performance in a hall. Whichever FR curve you choose, you're wrong, it's a no win concept. If you want to reproduce the tone accurately, you must also reproduce the reverberant field accurately, a task which for Bose 901 in a conventional sound system like any other speaker is hopeless.

Link to comment
Share on other sites

Interesting about the fiberglass. Are the drivers 72 ohms wired in parallel or 8 ohms with three in series and each series combination in parallel with the others to get back to 8 ohms?

The absorption of high frequencies in a concert hall is a dynamic event. The direct field has the full high frequency compliment that each musical intrument radiates in the direction of the listener. As the sound reverberates, the high frequency component is absorbed by both humidity in the air and the boundary materials of the hall selectively, that is high frequencies are absorbed faster than middle and low frequencies. The RT or reverberation time which is the time required for sound levels to fall by a factor of one million, that is 60 db in a typical modern concert hall is about 2 seconds at 1 khz, and about 1.2 seconds at 8 khz, the limits of the data I have. In older halls it is even longer, as much as 3 to 3.5 seconds at mid and low frequencies. This makes newer halls and renovated halls less reverberant increasing their clarity at the expense of reverberation (all other things being equal.) Because the spectral relationship of fundimentals and harmonics is a dynamic phenomenon which affects the perception of tonality, it is impossible for home sound systems using the current paradigm to reproduce the sound of musical instruments as they would be heard at a live performance in a hall. Whichever FR curve you choose, you're wrong, it's a no win concept. If you want to reproduce the tone accurately, you must also reproduce the reverberant field accurately, a task which for Bose 901 in a conventional sound system like any other speaker is hopeless.

The 801 I and II have 8-ohm drivers, measuring around 7.0 to 7.1 ohms dcr. according to the four spare units I measured. This would be a series-parallel wiring scheme.

As you know, low-bass output is a "pressure-volume" affiar, and being a sealed enclosure, the cone travel greatly increases with each lower octave. Equalized flat down to 30 Hz or so would mean that the cones would have to have pretty good excursion capability. When I remove the failed driver in my 901 I will dissect it -- if it is not a wiring problem -- to see the exact voice-coil overhang. From what I can tell so far, however, each driver appears to have approximately a 10mm excursion, or an Xmax of 5mm, and by the time you multiply all nine drivers, the total radiating area is somewhere between a 12-inch and 15-inch woofer, therefore the air-moving capability is not insignificant. By contrast, the original AR-1W/AR-3 woofer has a 15mm linear excursion (5/8-inch), with a 7.5 mm Xmax.

You lost me on your absorption discussion, so I'll take your word for it that the 901 cannot reproduce the reverberant field accurately. At least it gives a decent "illusion" of the symphony-hall soundfield.

--Tom Tyson

Link to comment
Share on other sites

"You lost me on your absorption discussion, so I'll take your word for it that the 901 cannot reproduce the reverberant field accurately. At least it gives a decent "illusion" of the symphony-hall soundfield."

When you sit in the audience of a concert hall, the first sound you hear from each note of each instrument arrives with the most relative high frequency content it will have. This is the sharp attack at the leading edge of a piano key being struck. This is the highest harmonics of violins. Then you hear thousands of echoes of the same sound arriving later from different directions each with a different relative spectral balance. The later the echoes from that note, the relatively less high frequency content the echo will have relative to its lower harmonics and fundimentals. That first sound is only 11% of the total sound of each note if you are sitting 19 feet from the stage of Boston Symphony Hall and if you look at Bose's graph in his white paper, the further back you sit the lower the percentage gets. Before the echos die out they can travel half a mile or more bouncing around inside the hall (2+ seconds at 1100 fps.) This means that the timbre of the sound changes as it dies out. This affects your perception of it. Without those echoes, you will hear only one spectral balance, a slice of it at one instant in its decay. If the speaker has a rolled off high end to achieve an "average" of the spectral balance over time, its percussive attacks and the bite of strings, brass, pianos and other hf content resulting from the leading attack of these instruments will be lost and the sound will be somewhat muffled, musted, indistinct. If you design for flat response of the initial attack, you will hear what those instruments would sound like in a small room where echoes die out quickly and this direct sound constituties most of what you hear. But there are other penalties to pay for listening to large groups like symphony orchestras heard that way. I can't go into them now but it should be evident that they will not give the same subjective tonality you will hear from them at a live concert.

It was instructive to listen to rehearsals of a local college symphony orchestra in a practice room on a Tuesday night before a concert, a "dress rehearsal" in the empty concert hall on the following Thrusday night, and at a live performance performance in a packed house on Friday or Saturday night. All were different. (My persnal preference was for the empty hall as I am what you might call a reverb freak.) I've was exposed to high reverberant sound fields by conincidence within a couple of days of being born and during most of my infancy and have studied them, modeled them, and experimented with them for the last 35 years. To this day I can't walk under a concrete overpass without yelling to hear the echoes of my own voice.

Link to comment
Share on other sites

  • 4 weeks later...
Guest boseguy

Tom Tyson said,

"Quite frankly, if the speaker has strong and smooth output up to 13-15 kHz, most people will find that to be realistic and fulfilling, and the lack of the top-most octave (not heard by the bulk of listeners anyway) is probably more of an academic issue than a true acoustical problem."

After I left Bose and was at another well-known speaker company, we conducted extensive listening tests and evaluations of what HF limit constituted "high fidelity." There were several products we were investigating that for budgetary/design/space reasons, a separate tweeter was not practical.

The "magic" frequency turned out to be around 13kHz. The research/listening tests were absolutely unambiguous about this. A device that extended to 11k was clearly not 'high fidelity' to, say, a 30 year-old listener with normal hearing on normal, high-quality program material. 12k was better, but at 13-13.5k, the preferences and ability to reliably identify between that and 20k (full-range) devolved into statistical insignificance.

Soundminded's contention that top-octave reproduction is necessary for convincing life-like sound is absolutely correct, since the "top octave" extends down to 10kHz.

boseguy

Link to comment
Share on other sites

  • 4 years later...
  • 2 weeks later...

Carl,

Simply Speakers has a You Tube presentation video where they walk you through the removal of the grills for the early 901 series models and also include info about the later series 5 and 6 models. It is quite informative so you might check it out.

http://www.youtube.com/watch?v=dbGfFw650gQ

I found it by googling Bose 901 Speaker Foam Edge Replacement and Repair.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...