AR Speaker Protection from DC Current

[ben76]

xPat,

Your diagram is nicely drawn and very clear. However, as you suspected, it is not correct. You should not have any wire connecting the negative terminals of the capacitors to each other. This wire makes the capacitors sit on the cable like a bird on a power-line :-)

Below is a corrected diagram:

Indeed, it looks like this stuff is new to you. Please ask a friend to help you, or do not do this at all unless you are certain that you can do this without making any mistakes.

[xpat]

Hi Ben,

Thanks for the review and correction. I can follow the corrected diagram. I'll order the caps and test on an old amp I'm ready to pitch or give away. I assume I could replace the 4700uf caps with 2 10,000 uf caps and wire as shown in the corrected diagram?

Regards,

xpat

[ben76]

Yes you can obviously use two 10,000uF capacitors.

Make sure you order 4 identical capacitors. Somewhat like batteries, it is best not to mix makes and models because this breaks the symmetry of the circuit.

Good Luck,

Ben

[dynaco_dan]

Hi there I am having difficulties trying to do any line spacing when I write here now. I would suggest that the writer of the incorrect drawing re-edit his write-up and delete the original drawing so as there cannot be a dangerous drawing still downloadable.

[xpat]

Ben, Thanks for the feedback - it is appreciated. Dynaco Dan, I deleted the incorrect drawing.

Regards,

xpat

[tysontom]

How about in-line fuses for protection? Didn't AR recommend a slow-blow Fusetron for the AR3a and 10Pi?

Hi, I agree with Gerry S on this one: probably the best protection against dc-offset failure is to fuse the speaker. Usually a slow-blow fuse of 2 amps (particularly the Buss FNM-Series fuses work well) will blow long before a voice coil will be set afire. The voice-coil bell wire used in the AR 12-woofer can probably sustain much more than this before it begins to overheat; however, the fuse will not protect the woofer against a short-term burst of energy that could cause other damage.

The AR Amplifier and AR Receiver both had output-circuit fuses for just this purpose, and these fuses came in handy. The AR amplifiers frequently developed bias problems and could inadvertantly dump the full dc rail voltage right into the output which would go right into the woofer's voice coil, typically immobilizing it and causing the coil to warp the former and freeze the cone in place. This was common in certain occurrences. Many transformerless amps and receivers from the 70s through the 80s did indeed have no output protection, thus could get into trouble if overloaded, etc.

If you were so lucky as to own a McIntosh amplifier, the output circuit is impedance-matched through an auto-transformer to specific output impedance values, and of course, those amps would never have any offset because a transformer will not pass dc under any circumstances. Because of the stability of these amps, the likelihood of failure of any kind with McIntosh was low, but never would the amp try to catch a speaker on fire.

The worst outcome is that a speaker can catch on fire with high-power dc offset current. The woofer voice coil simply gets red-hot and bursts into flame, then usually catching the cone material on fire and so forth.

Fuse the speaker! Better safe than sorry.

—Tom Tyson

[xpat]

Hi Tysontom,

Thanks for the information. I searched around my workshop and found (2) FNM2 fuses. I already have another set in the circuit to the AR-9s. However, my AR-5 speakers are not fuse protected. I'll probably order a couple FNM 8/10 fuses and open fuse blocks for the AR5s. I'm still going to put the caps in the circuit after the holidays just to see what happens.

xpat

[Gerry S]

I'm thinking the FNM slow-blow fuse approach is the simplest and least sonically intrusive. Those fuses are relatively huge and would require LOTS of "instantaneous current" to blow (as in catastrophic amplifier failure).

The trick is finding the correct value offering protection without blowing on musical peaks. These fuses can add DCR in series with the woofers if heating is gradual and continuous (gradually altering the bass response). But output levels would be such that your ears may bleed before you would notice it, or the woofers threatened.

[tysontom]

Gerry,

AR printed numerous guides showing the correct values to do just what you suggest: protection while not sonically intruding on the speaker's performance. I think we have some of these documents in the library section. The Buss dual-element FNM-series worked well for this purpose and was basically a motor-start-type fuse that allowed large short-term peaks while not allowing sustained high-current signals to pass to the speaker. These fuses would definitely protect against excessive or destructive dc offset (which would be sustained high-voltage, high-current output to a speaker). These fuses would generally protect against over-driving a speaker with a super-powerful amplifier or an under-powered amp that was clipping (saturation) badly. It would allow large momentary peaks to pass without blowing a fuse. An example that AR gives is the AR-7 being protected ty an FNM-6/10 fuse. This arrangement could pass peaks greater than 1500 watts which hava a duration of only 1/10th of a second. This is another reason why we always say that under-powered amps usually cuase more damage than high-powered versions, especially into low-sensitivity speakers such as AR speakers.

What the FNM fuse (or any fuse) would not do is protect a speaker from an instantaneous burst of high voltage, such as from the wall-circuit mains or from a direct lightning strike, etc. The fuse would probably blow, but not fast enough to keep the woofer from jumping out of the gap. Note, too, that repeated, excessive cases of fuse-blowing due to an amplifier's overloading or over-powering a speaker will usually cause cumulaive damage to voice coils.

--Tom Tyson

[dynaco_dan]

Hi there

Scroll down to "Other speakers, etc" and read the near top topic, "fuses for speakers fast + slow", and you will see most all fusing suggestions available here.

[Gerry S]

Hi there

Scroll down to "Other speakers, etc" and read the near top topic, "fuses for speakers fast + slow", and you will see most all fusing suggestions available here.

Thanks, Tom Tyson and Dynaco Dan. Brings back VERY distant memories. The published "protection data" given by AR was so detailed for a consumer product...never been equaled since to the best of my knowledge.

A zillion decades ago, I did order those "special" FNM fuses directly from AR for my AR 10pi's . I was surprised how physically large the fuse (and the fuse holder) was compared to a "regular" glass-type fuse. I permanently wired them into my speaker harness: 14 gauge Romex (in lieu of Monster Cable) with banana plugs on both ends. The harness was pretty unwieldy and ugly looking, especially when contrasted to the walnut veener finish of the speaker cabinets. I never blew a driver (OR the fuse as best as I can recall). And, I drove them pretty hard with my Dunlap-Clark Dreadnaught 1000. So, the ugliness was worth it.

If I were to get these fuses (and holders) today, I would probably have to go to an industrial electronic parts store; Radio Shack probably wouldn't stock them.

[dynaco_dan]

Hi there

As I've discovered, it is near impossible to find the FNM Buss Fusetron and equal FLM Littlefuse fuses and open fuse holders at retail.

Fuses can be found at surplus sites once in a while but usually only one or two values and the open fuse holder is rarer than a chickens tooth.

The last price I saw in an old Electro Sonic catalog here was $7.00+.

I have seen $2.00 and even less at surplus web sites for FNM 2 fuses.

Mind you there was a $10 - $15.00 minimum sale with a $15.00 S&H charge.

With Dynaco, Heathkit, Newer Advent's and others using either glass slow blow fuses or fast blow fuses we should start converting to an equal or lower value fuse.

Open style Littlefuse fuse holders for the glass fuses run around $2.00 and the fuses are around 5 for $3.00 for 1 amp fast blow.

addendum

If you are going to use fast blow fuses, a good safe starting size is 1 amp.

This will offer you the approximate equivalent to power handling of Dynaco A-10 ( also 3/4 amp ), A-25 and A-35 speakers, as recommended by Dynaco.

If you advance to 1 1/2 amp fast blow you would then roughly equal the Dynaco A-50 in power handling.

If you advance even higher to 3 amps you now have the Heathkit AS-103/103A/AR-3A equivalent output capability.

AR recommended a 1 1/4 amp FNM slow blow fuse for the AR-3A speaker system.

We cannot assume, at this time at least, that we can use a self drawn graph to compare a fast blow with the FNM/FLM fuses.

A glass slow blow fuse is not the equivalent of the FNM dual element slow blow fuse, so should not be used, unless further feedback proves me wrong.

With the 1 amp fast blow you will likely blow many fuses if you listen to loud or other rock music, but you can certainly obtain a reasonable sound level.

They will at least save your drivers, especially your tweeters, so spare fuses are always a great idea.

If you wish to add to the fusing topic please do so in the "OTHER" topic, please.

One can arrive at a close equivalent to the FNM/FLM fuses, but always to a safer and lower level.

[xpat]

Hi All,

FYI, I found a place that has the FMN fuses and holders at a reasonable price. The fuses and holders are en-route and should arrive today. Here is the link to the supplier:

http://www.zorotools.com/g/FNM%20Series%20Fuses/00075501/

Regards,

xpat

[LouB]

Hi All,

Guess I'm reviving a dormant thread, but here's the situation.

In another thread I was describing my rework on a pair of AR-3a.

As noted in the thread, I was planning to set them up with a rebuilt Dynaco ST-150.

Still will be doing that but have just acquired a Crown DC-300.

The 300 will be subject to a full rebuild also.

I am aware that there have been instances of failure mode sending full rail power to the speakers.

I want to protect the 3a's from certain doom and so wanted to create a redundant protection device.

My thought would be to use 2 amp slow blow fuse for wattage protection and a pair of 10,000 uf, 100V aluminum electrolytics positive to positive for DC blockage.

Being 4 ohm nominal, what would the consensus be as to whether the 5000 uf effective capacitance would be adequate for the oddities of the 3a impedance. 

Also, would the 2 amp fuses allow more than the 3a's will tolerate long term?

 

Thanks,

 

Lou

[owlsplace]

1 hour ago, LouB said:

...As noted in the thread, I was planning to set them up with a rebuilt Dynaco ST-150.

Still will be doing that but have just acquired a Crown DC-300.

The 300 will be subject to a full rebuild also.

I am aware that there have been instances of failure mode sending full rail power to the speakers.

I want to protect the 3a's from certain doom and so wanted to create a redundant protection device.

My thought would be to use 2 amp slow blow fuse for wattage protection and a pair of 10,000 uf, 100V aluminum electrolytics positive to positive for DC blockage.

Being 4 ohm nominal, what would the consensus be as to whether the 5000 uf effective capacitance would be adequate for the oddities of the 3a impedance. 

Also, would the 2 amp fuses allow more than the 3a's will tolerate long term?

Lou,

Here is what the voice coil of an AR 10" woofer looks like after receiving full rail voltage from a failed amp:

The possibility of a house fire seems plausible in this scenario.

I contemplated using the cap set-up in the LST's but eventually decided to just use AR's original fusing scenario.

I think this is Carl's post from AK:

Quote

 

This is what the LST owner's manual has to say about the large caps:
"A large capacitor in series with the autotransformer prevents excessive amplifier loading at subaudible frequencies where the transformer's inductive reactance decreases".

 

Gordon's post about extending the bass response by reducing the damping factor at low frequencies is interesting and I may have to test it out for myself. You will be adding the esr of the capacitors in series with the speakers so they will be dropping power. Here is the first version of the LST crossover:

I have some large power supply caps that can use to satisfy my curiosity about bass extension.

Here is Heathkit's fusing arrangement for their version of the AR-3a:

Roger

[LouB]

Ooooohhh, I like the Heath solution.

I'm not planning to sell the 3a's ever, so modifying them doesn't bother me.

And I may replace the tweets at some point with the HiVi subs anyway.

So the Fuses sound like an elegant solution to the overpowering issue.

Now I need to find some fuseholders with a longer than average thread length for the tightening nut.

Just not satisfied that this would protect against errant DC.

But it's certainly a good start.

Thanks for the idea,

Lou

[Carlspeak]

Capacitors will not pass DC current. Thus putting a very large value cap in series with the xover will block DC. The very large value cap will not induce any unwanted crossover slopes to affect the sound.

[LouB]

Thanks Carl,

I'm a bit wary with the 3a's and their weird impedance and the difficulty some amplifiers have with them.

The DC-300 is essentially an op amp capable of passing DC as a feature of it's design and if it fails...... Pffft!

[genek]

If you do a search for speaker protection, you'll find a number of products - most of them kits - for solid state devices that will detect amplifier faults and trip relays that will mute speakers.

You can get a stand-alone already built: http://www.parts-express.com/eminence-d-fend-sa300-loudspeaker-protection-unit--290-683

It's pro gear, so the price may seem steep, but if your speakers would cost 4x that or more to replace or over the years have become pure unobtainium, maybe not so much.

[JKent]

1 hour ago, genek said:

It's pro gear

Yeah--so no banana jacks. What kind of adaptors are needed with this?

There are cheap Chinese kits all over ebay. PE also sells a Velleman kit. Any good? http://www.parts-express.com/velleman-k4700u-2-channel-loudspeaker-protection-kit--320-264?utm_source=google&utm_medium=cpc&utm_campaign=pla

I use the fuses recommended by AR for the AR-3: FNM 8/10 A. They recommend 1-1/4A for the 3a (Why? I dunno I could probably use that size, since my 3s have 10Pi mids and Hi-Vi tweets so they're pretty robust but I've never had a fuse blow). Here's the document: http://www.classicspeakerpages.net/library/acoustic_research/special_sections/additional_ar_documents/fusing_ar_speaker_systems_1/

-Kent

 

[owlsplace]

6 hours ago, JKent said:

Velleman kit. Any good? 

IIRC when I was looking at these years ago the relays were too small for use with the Adcom 555II.

Roger

[LouB]

Given a choice between an active device and some dumb passives, my sensibility leans towards the passives.

Earlier in this thread, there was mention of  purported low end enhancement with the caps in place.

When I place the back to back 10,000uf 100V caps in line I'll post my observations as to whether this may be discernible (for me).

I'm trying to understand why that might be so, whether its that the bipolar capacitance in series changes the electrical reactance, causing a lowering of the resonant point.

If so, wouldn't that also affect the effective impedance?

 

Lou

[iso]

21 hours ago, LouB said:

If so, wouldn't that also affect the effective impedance?

Lou

This calculator and other Pronine online calculators are handy... you will see that reactance of 5000uF cap at 20 Hz is 1,6 ohms.... when you compare effect of series cap at 42 Hz, reactance of 5000 uF cap is similar to 2,85 mH AR3a woofer series coil ie 0,75 ohms... this has been discussed earlier here

 

Kimmo

http://www.pronine.ca/capimp.htm

[LouB]

On 12/10/2016 at 7:07 AM, iso said:

This calculator and other Pronine online calculators are handy... you will see that reactance of 5000uF cap at 20 Hz is 1,6 ohms.... when you compare effect of series cap at 42 Hz, reactance of 5000 uF cap is similar to 2,85 mH AR3a woofer series coil ie 0,75 ohms... this has been discussed earlier here

 

Kimmo

http://www.pronine.ca/capimp.htm

Very handy calculator!

That's just what I was needing.

I connected the 10,000uf back to back DC blocker caps to each other with a pull tie connected old screw terminals and hot melted the whole mess together. I then connected them to the 2Ax's.

I disconnected the Scott 477, and connected the Dynaco ST-150 along with the PAT-4A.

Both of these just had all the electrolytic caps replaced, power supply as well as signal path.

I didn't notice any apparent bass extension or enhancement but was happy to note that when the recently recapped ST-150 developed a cranky condition with errant semiconductors it indeed did exactly what I had intended. 

The unit developed intermittent DC pops and rustling sounds much like one may hear during shut down of an amplifier.

A trip to tri-state electronics for some PNP type 2N4889 should take care of the problem.

I didn't have any fuses installed, so can't say whether the AR recommended fuses would have opened with the DC pops.

Meanwhile, The Scott 477 ( old reliable ) is back in place.

Once again, with or without the caps in place, I don't notice with my old ears any significant difference between the caps or not.

Thanks for everyone's input on this!

 

Lou

 

[owlsplace]

29 minutes ago, LouB said:

... Once again, with or without the caps in place, I don't notice with my old ears any significant difference between the caps or not.

Good to know.