Speakerplans.com Homepage
Forum Home Forum Home > General > General Forum
  New Posts New Posts RSS Feed - Amp for running 6x Oberton XB1500
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

Amp for running 6x Oberton XB1500

 Post Reply Post Reply Page  <1 23456 7>
Author
Message
njw View Drop Down
Old Croc
Old Croc
Avatar

Joined: 26 March 2010
Location: S. Wales
Status: Offline
Points: 2562
Post Options Post Options   Thanks (0) Thanks(0)   Quote njw Quote  Post ReplyReply Direct Link To This Post Posted: 13 October 2018 at 10:35pm
Originally posted by ceharden ceharden wrote:

As per the other Chris's comment, in the absence of very expensive power measurement and/or processing, the best way to avoid blowing your system up is to use your eyes, ears and common sense.

I use clip limiters set slightly above the maximum output of the amplifier so it can achieve maximum Voltage swing but stops it the signal turning into a complete square wave!  However ultimately the limiter is me.  If I'm just tickling the limit lights occasionally then I take that to be the system running at maximum level. If I start seeing more than a brief flash every few seconds then it's time to turn it down.  The only difference is if I do leave a system unattended then I'll normally wind the limiters back a few dB to a more conservative level and probably increase the release time too.

As for the question of how much headroom, well it depends of course!  Mainly on the style of music.  I do a lot of live work and for that I like lots of headroom because of the high crest factor, especially on mids and highs.



 I'm much the same, I do mostly dj nights but always run a desk between the dj mixer and lms and keep on the eye on the levels as the night goes on with maximim level being the point at which I'm getting the odd flicker of the limiter lights. If I need to leave the system the levels get pulled back a few db and the lid goes back on the desk! Smile
Back to Top
gen0me View Drop Down
Young Croc
Young Croc
Avatar

Joined: 20 February 2016
Location: UK
Status: Offline
Points: 999
Post Options Post Options   Thanks (0) Thanks(0)   Quote gen0me Quote  Post ReplyReply Direct Link To This Post Posted: 14 October 2018 at 12:27am
Originally posted by Jo bg Jo bg wrote:

-thermal limiter , very attack long times, maybe 5 6 seconds under 100Hz, kicks in only when the threshold is reached for a long time, so only with high duty cycle/low crest factor  program.

It wont work. After bass ends limiter will reset and count attack again. It may not turn on at all. Also limiter turning on after few second  of bass would be annoying. Device has to have memory.

What are the maximum temperatures you see on powersofts?
How does it change with copper VC, aluminium and CCAW?


Edited by gen0me - 14 October 2018 at 12:28am
I appreciate every like :)) https//www.facebook.com/genomesoundsystems
Mixes: https://www.mixcloud.com/gen-ome/
Back to Top
gen0me View Drop Down
Young Croc
Young Croc
Avatar

Joined: 20 February 2016
Location: UK
Status: Offline
Points: 999
Post Options Post Options   Thanks (0) Thanks(0)   Quote gen0me Quote  Post ReplyReply Direct Link To This Post Posted: 14 October 2018 at 12:32pm
How powersofts know ωL part of impedance measuring music signal?
I appreciate every like :)) https//www.facebook.com/genomesoundsystems
Mixes: https://www.mixcloud.com/gen-ome/
Back to Top
gen0me View Drop Down
Young Croc
Young Croc
Avatar

Joined: 20 February 2016
Location: UK
Status: Offline
Points: 999
Post Options Post Options   Thanks (0) Thanks(0)   Quote gen0me Quote  Post ReplyReply Direct Link To This Post Posted: 15 October 2018 at 5:37pm
Originally posted by toastyghost toastyghost wrote:

Your drivers will still be able to be burned by 300-600w LONG TERM power. The rating alongside a driver is measured in ‘easy conditions’ with band limited high crest factor pink noise. It is not a real representation of power handling apart from in maybe a well tuned reflex box.

Originally posted by junoprobelaunch junoprobelaunch wrote:

Didn't see that last bit just got confused as to how we got on the topic of of how much power kills your drivers again. Allow me to check my understanding though... so when you connect a driver to an amp output and run power into it (say the amp gives out 1500watts),the driver resists the very vast majority of that and will only see a few hundred Watts tops? I assume the crest factor plays a part in this but am I correct in saying the above? 

You can separate speaker impedance into
ωL + Re + ~0.4%*Re*ΔT

Of which first one is inductance which affected returns power to amp. Second and third is resistance which affected changes this electrical power into heat and sound. (0.4% [Ω/°C] is temperature coefficient for copper.). 


Edited by gen0me - 17 October 2018 at 7:27pm
I appreciate every like :)) https//www.facebook.com/genomesoundsystems
Mixes: https://www.mixcloud.com/gen-ome/
Back to Top
junoprobelaunch View Drop Down
Registered User
Registered User
Avatar

Joined: 01 April 2018
Location: Devon
Status: Offline
Points: 298
Post Options Post Options   Thanks (0) Thanks(0)   Quote junoprobelaunch Quote  Post ReplyReply Direct Link To This Post Posted: 15 October 2018 at 9:45pm
Sorry gen0me, you've completely lost me Confused Why is that formula needed ? As in what does it actually show? I can gather a rough idea from the context but I'd ideally like more than just a rough idea ! You've also used some symbols that I've simply never encountered before such as ωL, ΔT and while we're at it what is Vamp? (the voltage an amp gives out ??).

Hard not to feel like an imbecile but I am so desperate to fully understand this that despite feeling as such, I simply have to ask. TIA
Back to Top
toastyghost View Drop Down
The 10,000 Points Club
The 10,000 Points Club
Avatar

Joined: 09 January 2007
Location: Manchester
Status: Offline
Points: 10920
Post Options Post Options   Thanks (0) Thanks(0)   Quote toastyghost Quote  Post ReplyReply Direct Link To This Post Posted: 15 October 2018 at 10:37pm
Originally posted by Timebomb Timebomb wrote:

Originally posted by toastyghost toastyghost wrote:

Originally posted by csg csg wrote:

in rough terms, peak limiters will protect your drivers from physical damage ( pole plating, hitting X-mech etc), RMS limiters will protect from thermal damage.

using peak limiters alone will not stop you during voice coils, and if used poorly can actually significantly increase the long term average power sent to your drivers.

without top of the range limiting, you have to think old skool - good driver / amplifier matching and listening to your rig when it is running - if it sounds happy it usually is...


And a good manufacturer does testing to set peak, RMS and also thermal, Xmax and power limiters to handle all situations, ideally with frequency shaped filtering too so that it can vary with signal content like a real speaker does.

This is currently an area I’m doing a lot of work on Timebomb - been a long time since we met, perhaps we should have coffee soon?


Yeah bob in if your up this way, im normally about,  i will do some more refined presets for my systems when the new processors are ready but its looking like the new year now... 


Doing an install in Preston sometime in the next few weeks so might make the extra hop up. Will send you a message
Back to Top
odc04r View Drop Down
Old Croc
Old Croc
Avatar

Joined: 12 July 2006
Location: Sarfampton
Status: Offline
Points: 5483
Post Options Post Options   Thanks (1) Thanks(1)   Quote odc04r Quote  Post ReplyReply Direct Link To This Post Posted: 15 October 2018 at 10:48pm
Originally posted by junoprobelaunch junoprobelaunch wrote:

Sorry gen0me, you've completely lost me Confused Why is that formula needed ? As in what does it actually show? I can gather a rough idea from the context but I'd ideally like more than just a rough idea ! You've also used some symbols that I've simply never encountered before such as ωL, ΔT and while we're at it what is Vamp? (the voltage an amp gives out ??).

Hard not to feel like an imbecile but I am so desperate to fully understand this that despite feeling as such, I simply have to ask. TIA


He's writing equation salad really to express a pretty simple concept.

If you want to know the amount of power reflected to an amplifier then you need the phase angle of the load and a little bit of simple trigonometry.

When an enclosure presents a non-resistive load to an amplifier, i.e. the bandwidth around a resonance or when the voice coil inductance becomes significant compared to the DC resistance then a proportion of the power delivered at those frequencies is reflected back to the amplifier and must be dissipated as heat. This is on top of the waste heat related generated by an amplifier being not 100% efficient that is already being disposed of. In my opinion this is important and is one of the primary factors as to why certain amplifiers are more reliable than others at sub. Generally the resonances when driving such cabinets are low, the power levels high, reflected power proportion with the wrong cabinet load could be enough to tip a design over the edge. It also depends a lot of your musical content. So a lot of variables all bunched up together.

The problem is that unless you know what you are doing and spend a lot of time with measurement gear then you'll never really know the numbers, although quite a lot can be inferred from an impedance graph of your enclosures.

Frankly it is not worth worrying about in the real world unless you have a particular fascination with making the measurements. You'd also have to integrate them over time and frequency to get real power figures making it all rather complex (no pun intended). It is something that interests me and one day I may try and get real numbers, but it'll take years to get around to.

People have given good advice. Buy decent units, run the system sensibly, use your eyes and ears. Set the limiters to keep the amps out of clip, use a good gain structure. There really is not a lot to it imo.
Back to Top
junoprobelaunch View Drop Down
Registered User
Registered User
Avatar

Joined: 01 April 2018
Location: Devon
Status: Offline
Points: 298
Post Options Post Options   Thanks (0) Thanks(0)   Quote junoprobelaunch Quote  Post ReplyReply Direct Link To This Post Posted: 15 October 2018 at 11:02pm
Originally posted by odc04r odc04r wrote:

Originally posted by junoprobelaunch junoprobelaunch wrote:

Sorry gen0me, you've completely lost me Confused Why is that formula needed ? As in what does it actually show? I can gather a rough idea from the context but I'd ideally like more than just a rough idea ! You've also used some symbols that I've simply never encountered before such as ωL, ΔT and while we're at it what is Vamp? (the voltage an amp gives out ??).

Hard not to feel like an imbecile but I am so desperate to fully understand this that despite feeling as such, I simply have to ask. TIA


He's writing equation salad really to express a pretty simple concept.

If you want to know the amount of power reflected to an amplifier then you need the phase angle of the load and a little bit of simple trigonometry.

When an enclosure presents a non-resistive load to an amplifier, i.e. the bandwidth around a resonance or when the voice coil inductance becomes significant compared to the DC resistance then a proportion of the power delivered at those frequencies is reflected back to the amplifier and must be dissipated as heat. This is on top of the waste heat related generated by an amplifier being not 100% efficient that is already being disposed of. In my opinion this is important and is one of the primary factors as to why certain amplifiers are more reliable than others at sub. Generally the resonances when driving such cabinets are low, the power levels high, reflected power proportion with the wrong cabinet load could be enough to tip a design over the edge. It also depends a lot of your musical content. So a lot of variables all bunched up together.

The problem is that unless you know what you are doing and spend a lot of time with measurement gear then you'll never really know the numbers, although quite a lot can be inferred from an impedance graph of your enclosures.

Frankly it is not worth worrying about in the real world unless you have a particular fascination with making the measurements. You'd also have to integrate them over time and frequency to get real power figures making it all rather complex (no pun intended). It is something that interests me and one day I may try and get real numbers, but it'll take years to get around to.

People have given good advice. Buy decent units, run the system sensibly, use your eyes and ears. Set the limiters to keep the amps out of clip, use a good gain structure. There really is not a lot to it imo.


That was much more helpful! Don't mean to be ungrateful to Gen0me but i feel like the above needed to be read before I understood what you meant. Yes, when you put it like that there isn't alot too it but when phrases such as "burning out your voice coils" and "expensive recone bills" are used to describe a concept, I just had to make sure I understood it 110%. Thank you
Back to Top
gen0me View Drop Down
Young Croc
Young Croc
Avatar

Joined: 20 February 2016
Location: UK
Status: Offline
Points: 999
Post Options Post Options   Thanks (0) Thanks(0)   Quote gen0me Quote  Post ReplyReply Direct Link To This Post Posted: 15 October 2018 at 11:15pm
Originally posted by junoprobelaunch junoprobelaunch wrote:

Sorry gen0me, you've completely lost me Confused Why is that formula needed ? As in what does it actually show? I can gather a rough idea from the context but I'd ideally like more than just a rough idea ! You've also used some symbols that I've simply never encountered before such as ωL, ΔT and while we're at it what is Vamp? (??).

Hard not to feel like an imbecile but I am so desperate to fully understand this that despite feeling as such, I simply have to ask. TIA
Vamp is the voltage an amp gives out.
Δ = delta = difference.

This is to show how much power of an amp that was supposed to give kW is really being taken by the speaker just in standard working conditions.
ΔT is the difference of temperature. Lets say ts parameters were measured on ambient temperature speaker in 20°C. Than when voice coil heats up to 120°C its resistance will grow by ΔT* temperature coefficient. So for copper voice coil that has temperature coefficient ~ 0,4%=0.004 [Ω/°C] and ΔT=100°C the Resistance will grow to Re + 100*0.004Re= 1.4 Re

Forget it. I figured out different way.


Edited by gen0me - 17 October 2018 at 9:12pm
I appreciate every like :)) https//www.facebook.com/genomesoundsystems
Mixes: https://www.mixcloud.com/gen-ome/
Back to Top
odc04r View Drop Down
Old Croc
Old Croc
Avatar

Joined: 12 July 2006
Location: Sarfampton
Status: Offline
Points: 5483
Post Options Post Options   Thanks (0) Thanks(0)   Quote odc04r Quote  Post ReplyReply Direct Link To This Post Posted: 16 October 2018 at 7:25pm
But apart from a matter of interest this is irrelevant to setting limiters, unless you are limiting current. And even then it is irrelevant because as voice coils get hotter they draw less power from an amplifier for a given frequency given that voltage is constant. P=V^2/R. Set for voltage and power dissipation is in fact at a worst case when the voice coil is cold.

The question of what happens when a voicecoil gets hot is decided between the power that the coil is continually asked to dissipate and the cooling available to it which is often a function of driver displacement. I.e. it is complicated. Differential equations abound and it is linked to the frequency content of your playback.

Voicecoils draw less power when hot, this is known as power compression and is a good reason why it is better to push 200% more drivers 50% of their capabilities vs 100% of drivers to their limit for a happier result all round.

The split ratio of the power supplied to a voicecoil between heat generation and driver motion via interaction with the fixed magnetic field is another concept. Probably not going to help when the magnet gets hot though, especially if you go past the Curie point.

Unless you are talking about implementing signal limiting based on voicecoil temperature detection via resistance rise over nominal. This is something I thought about before but as always never got around to trying. I think the biggest problem will be that accurately detecting the shift in reistance dynamically in-situ considering how differently the enclosure impedance will be when getting hammered vs a room temperature measurement, would be an interesting study. I'm sure someone has done something like it before.
Back to Top
gen0me View Drop Down
Young Croc
Young Croc
Avatar

Joined: 20 February 2016
Location: UK
Status: Offline
Points: 999
Post Options Post Options   Thanks (0) Thanks(0)   Quote gen0me Quote  Post ReplyReply Direct Link To This Post Posted: 16 October 2018 at 9:28pm
On left reasonable reflex. On right reasonable horn. Horn could use speaker's excursion a bit more.

18SW115 has Re=5.3ohm which on 170°C will give 5.3*1.6=~8.5 ohm
Here are impedance charts after changing Re in ts parameters to 8.5ohm as effect of rising voice coil temperature from 20°C to 170°C.

Power on the speaker would be P=U*I*cos(θ)/2
So if the amp was on spec measured on resistive load than
P=U^2/(2*R)
U^2=2*P*R
Knowing Z from hornresp we can calculate I=U/Z
Power on the speaker will be: I*U*cosθ/2
Power=U^2*cosθ/2*Z=2*P*R*cosθ/2*Z=P*Rcosθ/Z

R is impedance on which amp was measured
P is amp power. I assume that above that power thd rises a lot as I am using this power to estimate amp max peak voltage. The above equation on power from current, voltage and angle is for sinusoids.
Z is impedance from hornresp
θ is angle between Voltage and current, taken from hornresp

Can also scale power through efficiency and get separately acoustical power and heat.
So for 2kW amp here are the charts:

Notice that 18SW115 has very low Re as for 8ohm driver.


Edited by gen0me - 17 October 2018 at 9:10pm
I appreciate every like :)) https//www.facebook.com/genomesoundsystems
Mixes: https://www.mixcloud.com/gen-ome/
Back to Top
odc04r View Drop Down
Old Croc
Old Croc
Avatar

Joined: 12 July 2006
Location: Sarfampton
Status: Offline
Points: 5483
Post Options Post Options   Thanks (0) Thanks(0)   Quote odc04r Quote  Post ReplyReply Direct Link To This Post Posted: 16 October 2018 at 9:38pm
This chap has had a go at what I was thinking more or less - http://www.diva-portal.org/smash/get/diva2:1015303/FULLTEXT01.pdf Interesting read. Uses an adaptive filter approach to get Z(f) via impulse response as opposed to straight V(f)/I(f) calculation. Author concludes that the system is not accurate enough vs direct thermocouple. Inaccuracies in the current measurement and the relatively small shift in DC resistance under heating are names as suspected culprits. I think there are probably ways around a lot of those problems if one was inclined.

Some interesting notes are how the woofer Fs shifts down under heating + large signal abuse, neodynium magnets also estimated to lose ~10% flux at 100C.

Back to Top
 Post Reply Post Reply Page  <1 23456 7>

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 12.06
Copyright ©2001-2023 Web Wiz Ltd.

This page was generated in 0.066 seconds.