[rsaavedra]

Quote:

Originally Posted by cotdt /img/forum/go_quote.gif That seems like a good idea to me, for amps that have trouble driving 4 ohm loads. But your experience with the Dynahi shows that it's fine with 4 ohm loads.

Yes indeed, but it's only that if a frequency where the Atoms have just 4 ohms is played and is kept sustained for a long time, I would feel slightly worrisome about the Dynahi driving such a low load in such sustained manner for long. With the resistor I would know the Dynahi will always have a load above 8 ohms, no matter what frequencies get played and for how long.

 

[amb]

Quote:

Originally Posted by rsaavedra /img/forum/go_quote.gif I wonder if it would be too bad an idea (sound quality-wise) to put a high watt resistor in series with the speaker. Just to bring up that minimum impedance to let's say 9-10 ohms (a resistor of 5 ohms would do), in order to minimize potential challenges to amp stability.

Yes, very bad. If the small resistance of a fuse is bad, then a resistor of several ohms is real bad. There is a reason why thick, heavy gauge wires are usually advocated for speaker use... to lower the impedance. To add a resistor would be the antithesis of that.

As I mentioned above, the resistor will cause power loss delivered to the speakers. Let's say you use a resistor that is the same value as the nominal impedance of the speaker, then if the speaker impedance was perfectly flat you would have added a voltage divider that loses half the output voltage of the amp. The result is that only 1/4 of the power would be delivered to the speaker. In reality, since the speaker's impedance is not flat but vary wildly with frequency, the voltage divider action is not constant with respect to frequency, so that resistor will introduce significant frequency response changes.

What's more, in a properly designed vented or sealed speaker, the enclosure volume (and vent design) are carefully tailored to the woofer for optimum performance (this is called "Thiele-Small alignment"). One of the three important parameters is the total system Q ("Qts"), which comprises of electrical damping ("Qes") and mechanical damping ("Qms"). The electrical damping portion assumes an amplifier output impedance of zero ohms (or very close to it). By inserting a resistor in series, you will cause a change to the total Q and alter the speaker's bass response from the designed target.

Usually the audible effect of inserting a resistance in series with the speaker (other than a loss of power) is the muddying of the bass.

Edit: I should also mention that if the speaker is a multi-way design, then the series resistance will also wreak havoc with the crossover network response, causing further frequency response anomalies. In short. resistance in the speaker wire, whether it's in the wire itself, a fuse, or gasp, a resistor, is bad, bad, bad.

 

[rsaavedra]

Quote:

Originally Posted by amb /img/forum/go_quote.gif Yes, very bad. If the small resistance of a fuse is bad, then a resistor of several ohms is real bad. There is a reason why thick, heavy gauge wires are usually advocated for speaker use... to lower the impedance. To add a resistor would be the antithesis of that. As I mentioned above, the resistor will cause power loss delivered to the speakers. Let's say you use a resistor that is the same value as the nominal impedance of the speaker, then if the speaker impedance was perfectly flat you would have added a voltage divider that loses half the output voltage of the amp. The result is that only 1/4 of the power would be delivered to the speaker. In reality, since the speaker's impedance is not flat but vary wildly with frequency, the voltage divider action is not constant with respect to frequency, so that resistor will introduce significant frequency response changes. What's more, in a properly designed vented or sealed speaker, the enclosure volume (and vent design) are carefully tailored to the woofer for optimum performance (this is called "Thiele-Small alignment"). One of the three important parameters is the total system Q ("Qts"), which comprises of electrical damping ("Qes") and mechanical damping ("Qms"). The electrical damping portion assumes an amplifier output impedance of zero ohms (or very close to it). By inserting a resistor in series, you will cause a change to the total Q and alter the speaker's bass response from the designed target. Usually the audible effect of inserting a resistance in series with the speaker (other than a loss of power) is the muddying of the bass. Edit: I should also mention that if the speaker is a multi-way design, then the series resistance will also wreak havoc with the crossover network response, causing further frequency response anomalies. In short. resistance in the speaker wire, whether it's in the wire itself, a fuse, or gasp, a resistor, is bad, bad, bad.

I am completely convinced on not using it

Thanks Amb!

 

[rsaavedra]

Parts for my amps-to-speakers switchbox gathered already:

Got that headphone plug too to build a better Dynahi-to-switchbox cable for speaker usage.

Had that metal box and got the plastic one, not sure which one I'll use yet. Edit: will use the plastic one. Easier to keep the binding posts isolated that way.

The switches have two-positions. Each switch will be use to send one of the channels from either the Marantz AV receiver or the Dynahi's outputs to the corresponding speaker output posts in the switchbox. Here's a simple diagram:

Edit: wiring diagram updated to reflect the actual connections the way I arranged them in the box. Colors in the diagram indicate channel (Red for Right), not the color of speaker posts (there are two posts [red and black] per channel.)

 

[rsaavedra]

Switchbox in the making, 50% ready. Went with the plastic box because it easily isolates the binding posts from each other without any problem.

The plastic box alone is very tough, but added those MDF walls inside for reinforced sturdy attachment of the binding posts:

 

[amb]

rsaavedra, I hope those switches you use are break-before-make. Otherwise, as you move the toggle there is a short moment when the two amps will be "seeing" each other and that's as good as a dead short, and could potentially cause damage.

 

[rsaavedra]

Quote:

Originally Posted by amb /img/forum/go_quote.gif rsaavedra, I hope those switches you use are break-before-make. Otherwise, as you move the toggle there is a short moment when the two amps will be "seeing" each other and that's as good as a dead short, and could potentially cause damage.

Thanks Amb and yes good point, I had checked that using the multimeter, and there is no electrical transmission between the two inputs of the switch when it gets toggled, so yes they are break-before-make.

But even so, I won't be really toggling these switches frequently at all, or even while both amps are up. They will be toggled when switching between HT and stereo listening only, so in general I'll switch them with both amps off.

The warning prevails though, I do understand if they were make-before-break and the amps were both on (maybe even just one of them on) when toggling the switch, there could certainly be damage.

Hopefully will finish it tomorrow. Will post some pics as soon as it is ready.

 

[rsaavedra]

Switchbox ready!!!!


Used 14 awg stranded copper wire. Soldered all the connections between wires and the switches, and soldered small circular connectors to the other ends of the wires for better contact with the binding posts:

Here's the final look with everything finally connected. Facing any of the binding posts panels, the left column of posts corresponds to left channel, right column to right channel. (The little red cable ties mark the right channels both in inputs and output.) Each switch controls the ouput channel it has in front of itself. Switches pointing to the right select the binding posts to their right as inputs; to the left select the inputs from the left:

The diagram updated to match the arrangement above:

Here's the look of the box closed:

Spaced the connectors exactly 3/4 of an inch apart (vertically) so as to accept these kinds of speaker connectors:

Pics already at the Build photos thread.

 

[Thelonious Monk]

very nice.

if you like the dynahi a lot for speakers, looks like you could kill two birds with one stone (in one box)!

 

[rsaavedra]

Bad news, just blew one of my power supplies driving the Paradigm Titans with my Dynahi.

I was driving them playing Tool's Lateralus, same I had done with the Atoms, volume knob at about 1-2 o'clock. It wasn't as loud as with the Atoms, seems the Titans are less sensitive than the Atoms, despite specs indicating the same sensitivity.

Played the whole Grudge song and sounded great. At the sweet spot (9-10 feet away from the speakers) I measured most of the time between 81-83 dB during this song.

Then I sent the volume to 3 o'clock and started playing song #8, Ticks and Leeches. The sound was great, measuring 82-85 dB. (85 dB on average is the volume I like to listen to.) The bass and overall loudness was going up with the song, but about half a minute into the song, the sound went off completely

I checked the amp and the blue LED was still on, but there was no sound in either channel.

The way I wired the LED immediately made me think one PS could be out, because I power the LED with just one of the power supplies.

Disconnected everything and tested, and yes, one power supply is working fine; the other is not working. Both were pretty warm, particularly the transformers, but not really hot to the touch, since I hadn't played the amp for too long. Nothing appears blown or burnt at all, no burning smell either. The fuse of my enclosure is ok since one of the supplies still works perfectly. Not sure if the power supplies themselves have their own fuse though.

I can't really test the boards without both power supplies working, so I'm wondering if there might be any damage to the amp boards? Any thoughts/suggestions?

 

[cotdt]

the power supply you use must be too weak. why not use the official dynahi power supply, one PSU for each amp board?

 

[rsaavedra]

Quote:

Originally Posted by cotdt /img/forum/go_quote.gif the power supply you use must be too weak. why not use the official dynahi power supply, one PSU for each amp board?

The official uses one PSU per board? Didn't know that. The ones I'm using are 28V, 2 Amps each, don't think they are weak. Since each Dynahi board needs both +30V and a -30V leads, these two power supplies are wired in series, with the ground being the middle point between them. Thought the official dynahi PSU connection scheme and V/current capabilities would be similar to these ones'.

 

[rsaavedra]

Actually correction, and GOOD NEWS!!!! The power supplies are ok!!!!

And the amp too!!!

Not sure if it went into some protection mode that is temporary, but the two power supplies are working perfectly fine again, and the amp is working perfectly too.

Won't power the Titans again with it though

 

[n_maher]

It is very possible that the power supplies have some sort of thermally activated fuse/protection device to prevent damage if the demand gets too high. Probably a wise choice to not use those speakers again without finding out exactly what caused the shutdown.

 

[rsaavedra]

Quote:

Originally Posted by n_maher /img/forum/go_quote.gif It is very possible that the power supplies have some sort of thermally activated fuse/protection device to prevent damage if the demand gets too high. Probably a wise choice to not use those speakers again without finding out exactly what caused the shutdown.

My guess is too much current demand. The beginning of that song, Ticks and Leeches, really grows steadily into louder and louder with strong percussion all the way till a climax. And the volume knob was at 3 o'clock.

Is it possible this protection mechanism works based on current instead of temperature? It seemed to me that I have seen the power supplies similarly warm before when driving the headphones (after a long time of playback however,) yet this shutting down had never happened before.