IRS Beta Servo Open Loop Gain Setting
- Thread starter Nikkoholic
- Start date
Nikkoholic
Active Member
If you do the following test you can prove it to yourself. Leave all your equipment off and turn on just the Servo unit, after about 15 seconds the green light goes on and you hear the relay (I suppose) click on. If you unplug the servo cables the green light will immediately drop out, but it will come back on even without reconnecting the cable.
Same test but this time start with the cables unplugged, the servo green light will come on after about 15 seconds; just like normal.
I suppose even though this is a feedback / loop system; the servo doesn't really know or care if there is actually a signal from the accelarometer. Infinity should have incorporated system check that looks for (ohms?) some signal indicating the feedback loop is intact and operational.
I'm not an electronics person (so potentially making a fool of myself) but this is how it appears to be reacting.
By the way I would have never pulled the cables from the live servo unit had I not been on the phone with a fellow AKer who had already performed this test; who was assuring me it was OK... and it was.
When I pulled my cables out the servo did not originally turn off/ reset, since it didn't think cables were plugged in due to the wiring being crossed up. Howver, it must detect something since when the cables are aligned and all is working, the servo does detect the cable being removed and does reset??? A better question left to a servo expert or someone other than myself.
I am enjoying my new sound (bass).
Whatever! Just asking questions.Making sure! No harm is there?
Nikko,Thank you for sharing your results with me and others.
Last edited:
Hmmm...
The servo-correction system senses woofer movement. If it's not connected, it would always effectively report zero movement. So, if the amp sent a signal to move forward +1mm and then back through zero to -1mm, and the sensor showed zero movement, the correction would be to tell the amp to move more, in both directions, say +2 to -2. When that didn't work, it would try to correct even more, say +3 to -3. Quickly, it would be sending maximum correction most of the time, with less applied correction as the signal neared zero. At zero, there'd be no correction. On bass peaks, it'd be correcting like crazy, to the max it could, and still be sensing zero movement.
Wouldn't that max correction result in the muddy hyper-bass effect you heard? Escalating corrections would make a mess, but a loud one.
My Betas are about to go back into use, driven by a McIntosh MC-402 on the top and MC-602 on the bottom. Last time I used them (way back), I had a Threshold S-550e (250 wpc) on the bottoms. Now given 600 wpc for the servo to go crazy with, I'll be doing the tap test and checking those cables. Great thread!!!
The servo-correction system senses woofer movement. If it's not connected, it would always effectively report zero movement. So, if the amp sent a signal to move forward +1mm and then back through zero to -1mm, and the sensor showed zero movement, the correction would be to tell the amp to move more, in both directions, say +2 to -2. When that didn't work, it would try to correct even more, say +3 to -3. Quickly, it would be sending maximum correction most of the time, with less applied correction as the signal neared zero. At zero, there'd be no correction. On bass peaks, it'd be correcting like crazy, to the max it could, and still be sensing zero movement.
Wouldn't that max correction result in the muddy hyper-bass effect you heard? Escalating corrections would make a mess, but a loud one.
My Betas are about to go back into use, driven by a McIntosh MC-402 on the top and MC-602 on the bottom. Last time I used them (way back), I had a Threshold S-550e (250 wpc) on the bottoms. Now given 600 wpc for the servo to go crazy with, I'll be doing the tap test and checking those cables. Great thread!!!
i was talking to Nikkoholic.
I'd guess that the accelerometer outputs some steady-state signal back to the servo circuit when it's connected but not moving, that way if the connection is disrupted the system stops monitoring and doesn't freak out like you expect.
I think the reason the woofers don't sound great without the servo is that the box size is not optimal and there is no passive crossover the lower the resonance (like in the Kappa 9 and IRS Delta). With the servo, though, it doesn't matter - the woofers are forced to do what the signal says.
I think the reason the woofers don't sound great without the servo is that the box size is not optimal and there is no passive crossover the lower the resonance (like in the Kappa 9 and IRS Delta). With the servo, though, it doesn't matter - the woofers are forced to do what the signal says.
cflutist
Member
Sure wish I had found this thread earlier.
Just had Mr. Watkins re-foam 4 woofers (including the Servo Control ones), and I replaced the other 4 woofers with new from Harman for my IRS Betas.
Triple checked all the wiring but could not figure out why the left woofer tower was oscillating (Watkins said it was acoustic feedback) while the right one was fine.
I still can't figure out why changing the open loop gain on the back of the Servo unit from 0dB to -6dB cured the problem. The manual talks about the different gains of different amps, but I didn't change amps. Using the same 2 Aragon amps as before. So why did they run okay at 0dB before I refoamed, but not after? Does anyone have an explanation?
I too think there is too much bass from the towers and normally have the Bass Level at -4 to -3dB, sometimes I turn the Bass Contour down too.
15Hz is pretty dang low (pedal tone of an organ) ... it makes my windows rattle
Just had Mr. Watkins re-foam 4 woofers (including the Servo Control ones), and I replaced the other 4 woofers with new from Harman for my IRS Betas.
Triple checked all the wiring but could not figure out why the left woofer tower was oscillating (Watkins said it was acoustic feedback) while the right one was fine.
I still can't figure out why changing the open loop gain on the back of the Servo unit from 0dB to -6dB cured the problem. The manual talks about the different gains of different amps, but I didn't change amps. Using the same 2 Aragon amps as before. So why did they run okay at 0dB before I refoamed, but not after? Does anyone have an explanation?
I too think there is too much bass from the towers and normally have the Bass Level at -4 to -3dB, sometimes I turn the Bass Contour down too.
15Hz is pretty dang low (pedal tone of an organ) ... it makes my windows rattle
Have you tried switching the channels? It could be that one of the two sensors isn't hooked up correctly now. On my system, when the towers howl, they both do. I don't think there should be a difference.
cflutist
Member
When you say "sensor", what sensor are you talking about. Are you talking about the sensors on the actual woofer?
I had been playing the right woofer tower just fine while I was waiting for Mr. Watkins to return the last two woofers. I had sent them to him in 2 batches so that if one set got lost, I would still have one working woofer tower. So I had been playing the right tower only (disconnected the left one) while I was waiting for him to return the last 2 woofers for the left tower.
So when we installed the last 2 woofers in the left tower, we heard that howling sound. To confirm it was the left tower, we disconnected the left tower, and the right one was just fine. Then we disconnected the right tower, plugged in the left tower, and it was howling at 0dB. So we isolated the problem to the left tower.
At first we thought maybe it had something to do with the fact that both towers were operational now (and with 2 new Harman woofers in each tower).
Then we plugged the right tower back (both speaker wire and accelerometer cable) and it the left one still howled at 0dB. Turned that open gain loop switch from 0 to -6dB and the problem went away. I had never played with that switch before.
I am absolutely sure that the wiring is correct as my speaker wires have red/black banana plugs on the end, Red = right, Black = left. Those and the accelerometer cables were not touched on the Servo Control Unit or on the amplifiers. If one looks closely, the right accelerometer cable has a red marking on both ends. It was as simple as unplugging and then plugging in the speaker wire and accelerometer cable on each woofer tower. Neither of these were changed at the source. We also checked the woofer wiring 3 times, both in me tagging the +/Red side before we disconnected them, and in matching up the numbers with the Infinity wiring diagram (all the woofer wires in each tower have numbers on the e.g. +/Right Control Woofer is #16, while -/Left is #15, Bottom +/Right woofer is #20 etc.
So I'm still stymied as to why the left woofer towers played at 0dB before, but howls now at 0dB. Changing the open loop gain from 0 to -6dB fixes the progblem but why would re-foaming woofers change that? And how do those "sensors" come into play here?
Mr. Watkins talked about accoustic feedback ... why did I not have that before but do now at 0dB?
Thanks,
Cheryl
p.s. I don't know much about electronics, but I am a software engineer by trade and am familiar with debugging techniques used to isolate problems. That's why this is driving me crazy ... because it doesn't make logical sense to me. I was hoping that some of you electronics techie/geeks here would have an explanation.
I had been playing the right woofer tower just fine while I was waiting for Mr. Watkins to return the last two woofers. I had sent them to him in 2 batches so that if one set got lost, I would still have one working woofer tower. So I had been playing the right tower only (disconnected the left one) while I was waiting for him to return the last 2 woofers for the left tower.
So when we installed the last 2 woofers in the left tower, we heard that howling sound. To confirm it was the left tower, we disconnected the left tower, and the right one was just fine. Then we disconnected the right tower, plugged in the left tower, and it was howling at 0dB. So we isolated the problem to the left tower.
At first we thought maybe it had something to do with the fact that both towers were operational now (and with 2 new Harman woofers in each tower).
Then we plugged the right tower back (both speaker wire and accelerometer cable) and it the left one still howled at 0dB. Turned that open gain loop switch from 0 to -6dB and the problem went away. I had never played with that switch before.
I am absolutely sure that the wiring is correct as my speaker wires have red/black banana plugs on the end, Red = right, Black = left. Those and the accelerometer cables were not touched on the Servo Control Unit or on the amplifiers. If one looks closely, the right accelerometer cable has a red marking on both ends. It was as simple as unplugging and then plugging in the speaker wire and accelerometer cable on each woofer tower. Neither of these were changed at the source. We also checked the woofer wiring 3 times, both in me tagging the +/Red side before we disconnected them, and in matching up the numbers with the Infinity wiring diagram (all the woofer wires in each tower have numbers on the e.g. +/Right Control Woofer is #16, while -/Left is #15, Bottom +/Right woofer is #20 etc.
So I'm still stymied as to why the left woofer towers played at 0dB before, but howls now at 0dB. Changing the open loop gain from 0 to -6dB fixes the progblem but why would re-foaming woofers change that? And how do those "sensors" come into play here?
Mr. Watkins talked about accoustic feedback ... why did I not have that before but do now at 0dB?
Thanks,
Cheryl
p.s. I don't know much about electronics, but I am a software engineer by trade and am familiar with debugging techniques used to isolate problems. That's why this is driving me crazy ... because it doesn't make logical sense to me. I was hoping that some of you electronics techie/geeks here would have an explanation.
cflutist
Member
But "what" exactly was probably close?
Sorry to be so dense, I'm just trying to understand what is going on.
I know that the Open Gain Loop at -6 dB solves the problem, but I'm just inquisitive on how/why things work they way they do. The manual talks about different gains on amplifiers, but I didn't change my amps, they are the same amps that I used before. I'm trying to figure out what was in a steady state e.g. the amps, the wiring, the acceloremter cable, and what has changed to cause this.
I'm guessing, but here goes:
You already know that each tower has one woofer that has a sensor (accelerometer). What if, in the course of his work, Mr. Watkins made some change to the sensor in the left tower? It could be as simple as connecting it correctly, where it was wrong before, or cleaning it so it's more sensitive.
I would take out both s-woofers, switch them, and see what happens. Before removing them, I'd examine the way the sensors are connected to the jacks, looking for switched polarity, etc.
It seems clear that your system and the drivers are wired correctly, so the sensor function and connections are all that's left.
You already know that each tower has one woofer that has a sensor (accelerometer). What if, in the course of his work, Mr. Watkins made some change to the sensor in the left tower? It could be as simple as connecting it correctly, where it was wrong before, or cleaning it so it's more sensitive.
I would take out both s-woofers, switch them, and see what happens. Before removing them, I'd examine the way the sensors are connected to the jacks, looking for switched polarity, etc.
It seems clear that your system and the drivers are wired correctly, so the sensor function and connections are all that's left.