Design that has 24 db/octive roll-off at 100 Hz

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
about the amps

I also have a pair of these amps and found this site while googling for more info about them.

From the way these amps were used I was expecting a peak somewhere in the low end to counter the effects of the alignment(large 15" woofer in a small sealed box), and it seems like you've found it.

So far by just looking at the back of the amp I can see that the mid/high signal does not go through the sub amp/pre-amp sections. The mid/high input on the front of the amp is just passed through to the back where there are spade terminals for connecting the mid/high passive crossover (which is a separate unit).

Also, the sub pre-amp board has no passive crossover components, but does have a few op-amps, so I assume it could be active. The sub amp board does have two small coils (about the size of a nickel) and two mylar caps. I think this could be some kind of low pass filter that I've read digital amps use for noise reduction purposes, they just seem way to small to be used as a speaker low-pass filter.

Here's hoping we can make these things work.

NickB
 
diyAudio Moderator Emeritus
Joined 2001
Jimmy154 said:
The sound output on my motherboard audio out is more or less linear.


20 - 17.45 - .424
25 - ??? - .434
32 - 19.68 - .443
40 - 19.64 - .447
50 - 19.26 - .450
64 - 18.44 - .452
80 - 17.92 - .454
100 - 17.06 - .454
125 - 17.29 - .455
160 - 14.30 - .455
200 - 11.14 - .455
250 - 7.68 - .455
320 - 3.26 - .455
400 - 1.055 - .454

The first reading is my the amp at the gain set to the maximum setup (computer volume was at 40%, don't know if that matters).

Did you remove the resistors you had hooked up and drive this without a load? That step is very important. There is a decent chance that this might be actually a 16 ohm amp and you will never know it if you take your maximum voltage measurements with an 8 ohm resistor hooked up. Either try a 16 ohm resistor, or even better, no resistor.

Why might it be a 16 ohm amp? Because at certain frequencies near bass cutoff, both sealed and ported system have a region of very high impedance, and a high impedance amp can deliver more power into that than a 4 ohm or 8 ohm amp can.



Jimmy154 said:
Got any idea about the x-over slope :confused:

You might want to take a couple of more measurements at 500 Hz and 600 Hz. Hate to keep sending you back like this.

125 Hz seems to be where the voltage starts dropping. In the first octave, from 125 Hz to 250 Hz, you drop about 7 dB. From 250 Hz to 400 Hz-about 2/3 of an octave-you drop 17 dB. That would be around 24 dB/octave, but you might want to take those readings at 500 and 600 Hz to get the full octave.
 
diyAudio Moderator Emeritus
Joined 2001
This is just a little 6.5" Peerless 833599 speaker in a 36 liter seled box.

The red line is the impedance. As you can see, there is a region of high impedance near cutoff that a high impedance amp can deliver more power into, therefore boosting output at 30 Hz and 40 Hz.
 

Attachments

  • peerless 833599 sealed.gif
    peerless 833599 sealed.gif
    21.3 KB · Views: 92
diyAudio Moderator Emeritus
Joined 2001
Same story for a 36 liter vented box with the same Peerless 833599 woofer. High impedance in the 20 and 40 Hz region allows extra boost if the amp delivering the power has high impedance.

So I am wondering what the amp rating of your amplifier is, plus what the voltage is with no load resistors.
 

Attachments

  • peelress 833599 vented.gif
    peelress 833599 vented.gif
    22.2 KB · Views: 91
20 - 17.8
25 - 25.4
32 - 19.8
40 - 19.7
50 - 19.3
64 - 18.5
80 - 18.0 or 20.15
100 - 17.06 or 19.2
125 - 17.3
160 - 14.3
200 - 11.14
250 - 7.66
320 - 3.65
400 - 1.218
500 - .341
600 - .109
640 - .076-.070

The readings seem to fluctuate at certain frequencies and at 80 and 100 I get 2 readings and 640. I did not use any resistors this time or the time before. The amp is rated at 500 watts. The readings match up well from the ones before except 320 and 400 Hz are higher because I used a different voltimeter this time, at least I think that's why. At 125 I have 17.3 volts, so if I had a 24 db/octive slope the volts should be like 1.08 at 500 Hz, no :confused:

The subs that I got to go with these amps are 4 ohms nominal. I was told by some one (forget who) that the amp was designed to power a 4 ohm nominal sub. I also have the subwoofer specs. They're at this place http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=3068912652&category=18804


NickB,

Where did you get your amp at, ESS?
 
The gain on the amp is set to max. I suspect that Kelticwizard told me to do this because the peak might disappear on some amps if you turn them all the way up. Never asked him though.

The peak comes back when you turn the amp down. I think the peak matches the subs I bought at that link I gave above anyway :cool:
 
diyAudio Moderator Emeritus
Joined 2001
Sreten:

Apparently the peak is made by a preamp circuit before the main amp circuit. It boost 30 Hz up as long as there is headroom before reaching full power. Once the 30 Hz peak reaches full voltage at 19.3V, it stays there until the other frequencies rise up to reach full power as well.

So 32 Hz reaches 19.3 volts when 80 Hz is only at 3.9 Volts. When 80 Hz is moved up to 5 volts, 32 Hz has to stay at 19.3 volts. When 80 Hz is moved up 15 volts, 32 Hz still must stay at 19.3 volts. And then 50 Hz finally moves up to 20 volts or so, 32 Hz still cannot rise above 19.3 volts, and the peak is finally flat.

Therefore, the peak disappears at full power, returns as the power is lowered. :)

So this looks like a 100 Watt/channel amp into 4 ohms with a big bass peak at 30 Hz or so.

Any chance to test it with a 4 or 5 ohm load?
 
What can I use for a 4 or 5 ohm load? I used to have a box for these subs, but I threw it away a long time ago :bawling:. I don't think I can use the subs free-air.

How do you know the amps are 100 watt :cannotbe:. Don't you have to know what the amps are to get power :confused:

If they're 100 watt I can get some smaller sub(s). That would be :apathic:.
 
diyAudio Moderator Emeritus
Joined 2001
Jimmy154 said:
What can I use for a 4 or 5 ohm load? I used to have a box for these subs, but I threw it away a long time ago :bawling:. I don't think I can use the subs free-air.

How do you know the amps are 100 watt :cannotbe:. Don't you have to know what the amps are to get power :confused:

If they're 100 watt I can get some smaller sub(s). That would be :apathic:.

Radio Shack has some 10 Watt 1.0 ohm resistors, 2 for 99 cents.
#271-131. Old catalog-maybe the proce went up a little.

I would put five resistors in series. Just don't keep the amp at full voltage very long. Crank the amp up for a couple of seconds while you measure the voltage, then turn it down. Let the resistors cool, if necessary. Do that for a few different frequencies below 100 Hz-that is all that is necessary.

How do I know the amp is 100 watts into 4 ohm?

Power = V²÷R

V=20 volts, (approx)
R=4 ohms or 8 ohms

For 4 ohms:
Power= 20²÷4 = 400÷4 = 100 watts

For 8 ohms
Power= 20²÷8 = 400÷8 = 50 watts.

5 ohms is close enough to check the 4 ohm spec. :)
 
diyAudio Moderator Emeritus
Joined 2001
The 100 watts is no longer a really powerful amp, by today's standards. It's not weak, but there are reasonably priced amps that are more powerful.

What bothers me is that bass boost of 15 dB or so. It really does not make the amp more powerful at 30 Hz, just pushes the power up quicker in that region.

So if you tailor your speaker to be 15 dB down at 30 Hz, For the amp to get full use of the boost, the output at 100 Hz would have to running on 16 watts.

Here is a page of subwoofer amps from Parts Express. These give you your choice of flat output or adjustable boost, and some are quite a bit more powerful than the amp you have.
http://www.partsexpress.com/webpage.cfm?webpage_id=3&SO=2&&DID=7&CATID=43&ObjectGroup_ID=505

Either that, or I would suggest opening the case and seeing if you can bypass the boost circuit. I would rather design a sub that went flat to 30 Hz or so, if my amp is only 100 watts. You really don't have that much power to boost at 30 Hz while giving big sound at 80 or 100 Hz.
 
kelticwizard said:
Sreten:

Apparently the peak is made by a preamp circuit before the main amp circuit. It boost 30 Hz up as long as there is headroom before reaching full power. Once the 30 Hz peak reaches full voltage at 19.3V, it stays there until the other frequencies rise up to reach full power as well.

So 32 Hz reaches 19.3 volts when 80 Hz is only at 3.9 Volts. When 80 Hz is moved up to 5 volts, 32 Hz has to stay at 19.3 volts. When 80 Hz is moved up 15 volts, 32 Hz still must stay at 19.3 volts. And then 50 Hz finally moves up to 20 volts or so, 32 Hz still cannot rise above 19.3 volts, and the peak is finally flat.

Therefore, the peak disappears at full power, returns as the power is lowered. :)

So this looks like a 100 Watt/channel amp into 4 ohms with a big bass peak at 30 Hz or so.

Any chance to test it with a 4 or 5 ohm load?


I'd suspect some complex manipulation of output impedance,
and the amplifier is simply hitting the end stops circa 30Hz.
Testing into various loads must be done to ascertain this.

The Sunfire amplifier I believe is a rail switching design and 18V
seems reasonable for one rail, the other option is a very low
impedance driver circa 1Ohm.

Does the sub have this panel and what is the setting ?

The amplifier shown has a nominal rating of 500W.

:) sreten.
 

Attachments

  • acoustic-research-ar1-speaker-rear-panel.jpg
    acoustic-research-ar1-speaker-rear-panel.jpg
    16.2 KB · Views: 159
well...

Thanks for the welcome guys. I also got these amps for the same EBAY source about 2 years ago. However, I'm pretty sure these put out a lot more than 100 watts, the torroid x-former on these things is huge. I'll try to get a picture of the internals uploaded tomorrow morning.
 
some info,

OK, I made a CD with some test tones and hooked up my digital meter straight to the amp. Here's what I recorded:

20 - 13.47
25 - 20.94
32 - 22.75
40 - 19.15
50 - 14.38
64 - 10.26
80 - 9.55
100 - 9.44
125 - 8.51
160 - 5.81
200 - 3.23
250 - 1.46
320 - 0.49
400 - 0.15
500 - 0.04
600 - 0.009
640 - 0.005

I recorded each test tone for 10 seconds, it took my meter about 3-4 seconds to settle on a steady value, which is the number I recorded. This was due to a 2 second gap between tracks.

Taking a look at the back of the amp I noticed a pair of Toshiba 2SA1301's and a pair of Toshiba 2SC3281's. I would think this amp would put out at least 300watts, no?

NickB
 
I think we're on our own prehaps NickB.

You should try what Kelticwizard suggested with the resistors to see how much output that amp is producing. I would do it myself, but I don't have the voltimeters I had anymore. Sreten said it might be a rail switching design, so this might not work, was that the point?

I did try hooking my sub that I have up to the amp and they did seem to push the sub pretty well in free-air. I could tell better how much power they produce if I still had that sealed box that I used to have designed for these subs.

Do you think you will see how much this amp can put out? I think I might try it with my analog voltimeter now that I give it some more thought.
 
diyAudio Moderator Emeritus
Joined 2001
Don't give up yet, guys. I think Sreten is very familar with the principles of this unique style amp. Let's see if he could come in here with some suggestions. As for me, I have essentially exhausted my meager knowledge of amps with this one.

Besides, I am very curious how you change an amp's output impedance for certain frequencies.

I was going to suggest measuring rail-to-rail voltage, just to see if that was far in excess of a 100 watt amp. But I think you just ought to wait a bit to see what Sreten has to say-apparently he knew the model well enough to post the picture of the plate amp.

If you don't get an answer in a couple of days, I would suggest starting a new thread about Carver sub amps. Somebody else has to be familiar with them-in fact, I think I might know at least one other poster.
 
diyAudio Moderator Emeritus
Joined 2001
By the way, are you two New Englanders eager to get going on this because you couldn't get to work today? Most things were closed around here, between the morning snow and the intense cold. :frosty:

It's 7º Fahrenheit scheduled to go down to -4º F tonight. That is -20º Celsius.

It's not scheduled to go above freezing here until next Thursday. and even then, it's only supposed to be 33º F.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.