testing a crossover


hi guys and girls i have a
problem with my 3 way  wilmslow-audio mirrage speakers , I am hoping you can help me with the problem,.
Lately my power amp went pop and i sent it away for repair,when i
received it back {repaired} it went pop again,on sending it back to
the repair technician ,he has come back to me ,and asked me to take
some ohms readings on the speakers, on testing the speakers , i run
these speakers using 2 power amps{Arcam Alpha 10s,using one for low
pass{bass} and the other for MID/HIGH,this is the amp which keeps
blowing,on the high /mid,on testing the terminal the bass{low pass}
read {left speaker]7.9 ohms ,right speaker bass 6.8ohms, But on the
left and right speakers hi/mid ,both read 0.0 ohms,{which we all know
is wrong}
BUT the strange thing is,when the blown amp was away,i only used one
power amp to run both speakers in bi wire config,and they worked
ok,all being a bit lack of bass and tightness,but still sound great.
So i stripped both speakers and checked the drivers,all 6 drivers
where good no shorts reading approx correct ohms for age,{Speakers
disconnected from x over}.can any one help with this,i would be so much in debt.
kind regards Mark
likklegerry
You have a dead short. They worked when connected normally because then the amp sees the sum total resistance including the 6.8-7.9 ohms. But connected bi-wire one amp sees only a dead short and so pop, there she goes.

The lack of bass and definition is because the dead short results in a total impedance much lower than 6.8. It is simply running out of gas and so what seems strange to you is perfectly normal and expected given the situation.

Now it seems something happened to create the dead short. This is what caused the pop and the first amp failure. It seems the drivers are putting out sound or you would have noticed that, and they measure fine. So probably the short is somewhere between the terminals and the drivers. That leaves wires and crossover.

Get in there and see. (I hope it is not one of these potted crossover situations....)
I agree with MC's dead short analysis. Are you b-amping using the passive crossovers inside the speakers, or are you using an active crossover at the inputs of each amplifier? If you are using the passive crossovers in the speakers, then there could be a short in the crossover. But, that doesn't explain why it works when you go back to driving with a single amp. Let me think, and I might have more suggestions once you get back to us.
millercarbon
You have a dead short. They worked when connected normally because then the amp sees the sum total resistance including the 6.8-7.9 ohms. But connected bi-wire one amp sees only a dead short and so pop, there she goes.
Nope. You're confusing bi-wire with biamplification. They are two different things.

In biwire configuration, the amplifier still sees the drivers' total resistance.
Okay so MC said bi-wire and not bi-amp, big deal, I knew what he meant. At this point we know there's a short somewhere. Could be shorted caps, but we need more info.
Where the heck is George at?

You use a single amp and it works with a little less bass. "lack of bass and tightness". NOT no bass. If all the drivers are working and you run the XO in parallel, the reading will go LOWER not higher. Parallel a pair of 8 ohm speakers you get 4 ohms.

Put your DMM on the post that reads ZERO and leave it there for a a while. You have to charge the cap (s) with YOUR little battery in the DMM, do you understand? A lot more caps in the band pass than the BASS OX.

On the bass side there is usually just a low pass, a resistor for 6 db first order maybe a cap TOO, for a second order 12db..

Get some correct data FIRST then figure out what up.. I’m not 100% sure until I know what the mids and highs read on a DMM.

Going from bi-amping to bi-wiring and a single amp tells me there is an AMP problem or the way it’s being wired. A stray single strand or some weird thing like that.. A close look and make sure there is no stray wires touching and you somehow make that happen when you hook that amp back in. You don’t want a strand from one amp touching another.

IF the same speaker cables are being used and you're just adding or removing an amp, it sure sounds like an amp issue to me.

Less swapping a PC for the one that keeps taking a dump.. You did swap PC right? What else is there?

Check the RCAs and XLRs cables for the amp that is acting up. Just swap the silly things.. make sure.

Regards
The formula for parallel resistance is 1/Rt= 1/R1+1/R2+1/R3....   
https://www.electronics-notes.com/articles/basic_concepts/resistance/resistors-in-series-parallel.ph...

Since he measured almost 8 ohms then the total resistance will be 1/8 ohm + 0 ohm or 0.125ohm.  

I think we all can agree that is a pretty low resistance and would suck the bass and dynamics out of most amps. 

What I don't understand is where is the short? And how does the driver manage to still get enough signal to make sound even though there is a dead short?
Got me MC. I think the repair guy missed something.. If it was to heavy a load it will pop a fuse not blow up. Must have one sorry protection circuit.

TWICE. I don’t think so.. It’s the amp or PC or RCA or something exclusive to THAT amp..

The input to the amp?

It should pop a fuse if it’s on the output side, like a speaker issues, usually.

All I know is neg to neg and pos to pos = 1/2 the two numbers added together. 8 ohm top + 6 ohm bottom = 14 ohms /2 = 7 ohms

Running neg to pos, to neg to pos, combines the two. 6 ohm bottom + 8 ohm top = 14 ohms.

8 -14 ohms loads on my valve amps are just perfect. They run a LOT cooler and clear as a bell all the way to 100% volume..

I’m likin’ the bass at 8-12 ohms better too, with BIG class d amps. They are 12K amps. 1700 x 2 all day long at 12 ohms bullet proof. Dirt cheap!

Get a Behringer NU12000 (NO DSP) move your other amp to the top. Be happy for 400.00 and 10 years.. I disconnect two of the 3 fans and put the last one on a switch. Perfect.. DEAD quiet.. or hook it to 10-12 18" subs.. It will push EVERY one to 120 db.

I'm thinkin' 4 new 21" Dayton's with 8 18" passives would be a good bass system .. :-) I like a room that can tenderize meat in a pinch.. :-)
    Wilmslow Audio built some fairly complex crossovers, for some of their systems, though I’m not familiar with the Mirage’s.                                                                                   A DMM can easily read impedances, through the inductors in a woofer’s series circuitry.     The capacitors in mid/tweet series circuits are another story.                              I could (easily) be wrong, but: I’d put my money on the speakers/crossovers being OK and something in the amp, or cables/connections: the culprit.
I have a love/hate relationship with puzzles like this. On the one hand I love em because it is nice to think they can be solved if only you know enough. On the other hand I hate em because you never can be sure you are getting the whole story. Like, one train leaves at 30mph, another at 40, how many painters does it take to earn $40?  

In this case I can't help wondering, how on Earth did two speakers go bad at the same time? In other words maybe zero ohms is a red herring. No trains left the station. Three buckets of paint, and the waitress earned a $40 tip?
The formula for parallel resistance is 1/Rt= 1/R1+1/R2+1/R3....  
https://www.electronics-notes.com/articles/basic_concepts/resistance/resistors-in-series-parallel.ph...

Since he measured almost 8 ohms then the total resistance will be 1/8 ohm + 0 ohm or 0.125ohm.  
If any one or more of the paralleled resistor is 0 ohm, it is a short circuit and the total resistance should be 0 ohm.

The correct formula for two resistors in parallel is Rt = r1 X r2 /(r1+r2)
The best way I know of to test a crossover is to generate an impedance graph. You can use DATS V2 (or is it now up to V3) or you can build a jig and use Room EQ Wizard’s app for this.

But... if you are bi-amping it is possible the two crossover sections are not properly separated. If the voltage in one section can touch the voltage in another, at all then you can cause the amps to fight each other. If the internal crossover is completely separated, including separate grounds internally then no problem.

OTOH, if the ground is not separated internally and 1 amp has reversed outputs or balanced outputs then you run into the problem above.
@likklegerry - Anymore info from you? Are you using internal passive crossovers?
Hi Guys,thankyou so much for all your help,but i still dont see a solution to this problem,in my opinion there is shorts in both xos,please see below-

             LEFT SPEAKER                                  /
 RIGHT SPEAKER
                speaker terminals with everything connected on both speakers

     MID/HIGH        0.0 ohms                             /
    MID/HIGH    0.0 ohms
       BASS           7.9 Ohms                             /
       bass          6.8Ohms

                                      DRIVER TEST DISCONNECTED FROM X OVER
          TWEETER 8 OHMS                                 /
    7.9 OHMS
              MID 6.3 OHMS                                   /
         6.2 OHMS
             BASS  6.1 OHMS                                /
       6.0 OHMS

     CONTINUITY TEST ON X-OVER WITH DRIVERS REMOVED USING DIODE FUNCTION
                                             between positive and
negative on x over
         HF                  000.2 {dead short}
         /       000.2 {dead short}
         MF                  000.5
              /       000.5
         LF                     OL.
                /        OL.
MID HIGH INPUT IP     OL.                                 /
                OL.
   BASS INPUT            OL.                                 /
                   OL.
                                                  OL. BEING NO CONTINUITY

                               TEST TESTED AT/ON X-OVER SPEAKERS CONNECTED OHMS

         HF 0.0002 OHMS                                      /
              0.002 OHMS
         MF 0.0001 OHMS                                      /
               2.0 OHMS
         LF   6.4 OHMS                                          /
                   6.3 OHMS
ive also now removed one lead from all the caps on one xo,and tested them with a cap meter,all caps are the correct uf ,within spec,can anyone tell me where to go from here,
thankyou again for all your kind help.regards Mark
i am using internal xo,which came with the speakers
all tests done with power amps disconnected.
the problem{ going pop{broken}only happens when the speakers are driven in bi-amp config,and its the amp which is driving the high/mid drivers that blows up .
when using only one amp,in bi-wire config,everything works,all be it the sounds quality is slightly off,but i would have thought that to be ok,that why i bi-amp,for better sound quality.
to add to this,i changed out the xo caps for better ones two years ago,and after fitting them ,i tested the input speaker terminals for the correct ohms reading and all was good,but something has changed.
regards Mark
i would also upload some photos, but as this is my first ever use on this site,im not sure how to do this, any help appreciated,
mark
I would like to help as I'm an avid bi/tri-amper. I would guess you've checked the cables and wiring to the amp and speakers? I don't know anything about the crossovers in the speakers, is it possible to draw a schematic or shoot some photos? I use an active crossover at the inputs of my amps to limit the frequency range to each driver, tri-amping. Nothing but wire between the voice coils and amp outputs. This configuration virtually eliminates intermodulation distortion, which is most offensive to the ears. I've been bi-amping for 44 years and tri-amping for 41 years. Might be as simple as bad caps in the crossovers you are using. If you would like, I could send a PM to you with my email and you could send me a schematic or photos of your speaker crossovers. I'm positive we can get to the problem without much effort. 
i would also upload some photos, but as this is my first ever use on this site,im not sure how to do this
Go to Virtual Systems page, use the "Create System" to upload photos.

https://systems.audiogon.com/
that all sounds fantastic Fiesta,ill get on that today,taking photos,ill try draw a schem.but its difficult, but ill do my best,thankyou so much for your help here,its so much appreciated,im sure its some thing simple in the xo,but i cant see it,but it strange how both have gone the same way,caps where all changed out 2 years ago,but ill test them all again.
kind regards
Mark
Thanks imhififan, I didn't know that either. One other thing I thought you should check on both amps, both channels. How much DC voltage is present at the output terminals?
im trying to upload photos,but it keep coming up with 
" Title has already been used on one of your systems."
I DONT UNDERSTAND WHY ?
REGARDS
MARK
OK,IVE MANAGED TO SAVE THE PICTURES,but i am now strugling to put them on this page,im so useless !!,
regards
mark
imhififan1,thankyou for that.
regards
mk
Post removed 
I just looked at all of your crossover photos. Don't take the following comments the wrong way, but this is my honest opinion. You should take those crossovers out and tri-amp with an active electronic crossover, please. You will truly be amazed at the improvement in sound you will get. Those are pretty cheap crossovers as the inductors are wound on ferrite and should be "air wound" on bobbins. The wire they are using for the inductors is Way to small and likely sucking allot of power from the drivers. This was a cost cutting decision when they were designed. No matter what you decide, I will help you get your speakers up and running again, bi-amped or tri-amped, whatever you decide. I'm going to take a close look at the measurements you posted the other day and all the pictures again. Will come back with questions and probably additional tests to perform. Did you ever measure the DC voltage at the amplifiers outputs with nothing connected and no signal input? Did you get the damaged amp back repaired yet? 
Did you change all of the capacitors yourself? The caps look like they are good quality.
The reason why the speakers work with a single amp and not when bi-amped is because the crossover is using different sets of inductors and capacitors. When the speakers are bi-amped you are using one set of inductors and capacitors, those are the caps or wiring that has the problem. The inductors are not the issue in my opinion. The wiring is suspect, if the caps test good. Think about what I've said thus far and tomorrow we'll get into the testing if you insist on repairing these crossovers. At this point, I don't think it's necessary for you to measure the DC at the amplifier outputs although that is always good info.
ill try draw a schem.but its difficult, but ill do my best
@ likklegerry,
If you can print-out the crossover pcb photo and add all the component position on it with wiring and post the picture, maybe we can figure out the schematic and go from there.

 https://www.wilmslowaudio.co.uk/ekmps/shops/wilmslow/images/crossover-pcb-452-p.jpg

an example:

https://ibb.co/3zyQRyG


OK,lets have a go with this,i have asked wilmlow for adivice ,schems and any help they can give me,but they just say send it in to us,and are not coming foreward with any other help,which is a shame,as these speakers were not cheap,for me anyway..
ive checked the AC voltage on both amps
amp driving mid/high which has popped twice, now repaired but running in another system {all good}
readings left channel -0.001v {AC]   right channel -0.001v {AC]
good amp[never popped} runs bass, left channel 0.002v [AC]
                                                            right chan.  0.002v AC
all other components turned off, except amps 
At this point i should add a little more history,after i got the amp repaired and plugged it back into the system,after about 3 hours it went pop again,so i sent it back to the technican,after ttwo days of him looking at it,he sent me an email asking me to checked my speakers,which is when i realised that was where the problem was,so i dont think it is either amps,i now run the repaired amp in another system which seems to be good,so i think the problem is in the crossovers,as all driver read the correct-ish ohms,but what i find very strange,is that both xo s have gone us,this is a puzzle to me.
I am totaly open to any options,but as a retired guy,money is a factor,but would love to find out the fault,but if its easier,i have another two power amps in boxes,but there are lower in power {Arcam Alpha 9p s}.
Another question ,is it not AC at the speaker terminals ?,im not sure!!
I have removed the caps from one crossover{or one lead} and they check out to be ok on all caps,
What do you think maybe  suspect on the wiring,or is it just my soldering ?.
ANYWAY I THANK YOU SO MUCH FOR ANY HELP 
regards Mark
PS if you can give me a day to work out a schem.that would be great.
Mk
Are you measuring the resistance across IP- and IP+ on the PCB? It looks to me like IP+ routes to two resistors (2.2Ω and 3.3Ω), there's lots of clearance between the tracks and no obvious short. So even if there were a problem further down the line you'd expect to see at least 1.3Ω across the inputs. It's rare for resistors to fail closed circuit so I wouldn't go looking there just yet.
The 2.2Ω is in series with the 12µ cap which will block DC, I can't work out what the 3.3Ω connects to under that big cap, but I still can't see how there'd be 0Ω across IP- & IP+. I'm sure I'm missing something.
I can see fibres on the PCB conductors but assume they're hot glue or stuffing material. Even so I'd start by cleaning up the copper connected to IP- & IP+ and measuring the resistance across those if you haven't done so already. I'd also check the multimeter by measuring a known value e.g. a resistor that's out of circuit. You wouldn't believe the number hours I've wasted going down a rabbit hole only to find I've measured something wrong or missed something obvious.

I'm sure you'll get there in the end with the help of some others here... good luck.
No NOT ac voltage, DC voltage. This is the offset voltage of the amp and I wanted to make sure they were low as THAT could be a cause of capacitor failure.
Well if you have lifted one lead of each cap and they test good, I would say it's a short on the PCB. Check all the resistors just to be safe, most resistors open or go high when they fail. In one picture I saw a couple capacitor leads that looked very close together. On another note, you said the amp worked for about 3 hours when you first got it back. So I'm assuming they sounded good when bi-amped before it went pop? I'm thinking there could be physical contact between a couple components lead wires. Reason, it work for a while. Maybe vibration caused the components to get physically close enough to short. What is the value of the cap connected to the midranges in the bi-amp configuration?
For some reason I can't access your crossover photos I saw yesterday. Anyone else have this problem?
thankyou all again for this help,i uploaded a schematic of the crossover wilmslow have now sent me,i will test and check all readings and offset dc today,then i will get back to you all.
so many thanks
regards
mark
Hi Guys,thankyou so much for all your help,but i still dont see a solution to this problem,in my opinion there is shorts in both xos,please see below-

LEFT SPEAKER / RIGHT SPEAKER
speaker terminals with everything connected on both speakers

MID/HIGH 0.0 ohms /
MID/HIGH 0.0 ohms
BASS 7.9 Ohms /
bass 6.8 Ohms
Please disconnect the wires from binding post and measure the MID/HIGH resistance again across IP+ and IP- to rule out the short is not from the binding posts.

DRIVER TEST DISCONNECTED FROM X OVER
TWEETER 8 OHMS / 7.9 OHMS
MID 6.3 OHMS / 6.2 OHMS
BASS 6.1 OHMS / 6.0 OHMS

CONTINUITY TEST ON X-OVER WITH DRIVERS REMOVED USING DIODE FUNCTION
between positive and negative on x over
HF 000.2 {dead short} / 000.2 {dead short}
MF 000.5 / 000.5
LF OL. / OL.
MID HIGH INPUT IP OL. / OL.
BASS INPUT OL. / OL.

OL. BEING NO CONTINUITY

The above measurement of MID HIGH INPUT IP looks OK,

TEST TESTED AT/ON X-OVER SPEAKERS CONNECTED OHMS
HF 0.0002 OHMS / 0.002 OHMS
MF 0.0001 OHMS / 2.0 OHMS
LF 6.4 OHMS / 6.3 OHMS

The HF measurement should equal to the DC resistance of the 0.22mH inductor.
The MF measurement should equal to the DC resistance of the 2.5mH inductor + 2.2Ω.

My suggestion is desolder all components from PCB, clean up the PCB, measure all components before re-install to PCB, leave some gap between resistor and capacitor to avoid heat damage those capacitors.

fiesta75 and imhififan are on the money with their advice... but I'd hold fire with the soldering iron until you've got a little more to go on.
IP+ and IP- should be open circuit at DC (OL), although you may need to leave the probes in place a few seconds while the caps charge.
If you're still getting 0Ω across IP+ & IP- then it's most likely a short somewhere between the input terminals and the copper pours that sit directly under IP+ & IP- on the PCB, you should be able to check this visually.
If that doesn't turn up anything then it might be time to reach for the iron.
pragmasi - I agree, rule out the simplest possible problems first. That's why asked about cables and connectors. Did we ever get an answer about that? I'm going to look at the schematic now. Good luck!
@fiesta75 and pragmasi,
IMO, the layout of components is not ideal, especially those capacitors that seat on top of resistors! since he is at it, I think it's time to fix that...
I just looked at the crossover too, not good. Would be very easy to create a disastrous mistake, not to mention the EMI and RFI problems. Heat is not a friend of caps as you know.
Yup, I don't disagree and that's good advice. I'm mainly interested to see what the cause is... it's a weird one.
I can guess, but that would be uncalled for and premature.
I can guess, but that would be uncalled for and premature
Is that in response to me?.. don't know what you mean. Only point I was trying to make is if you change lots of things at the same time sometimes it's difficult to know what the original problem was and so you learn nothing along the way.
pragmasi - Yes, sorry for being an SA. You are absolutely correct, one needs to take one step at a time. I'm waiting to see and help if possible.
The mid xover is a bandpass network and tweeter is high pass network which means if you measure the resistance, it should read "infinite resistance" or something very high.  It's interesting that both left and right channel all read 0ohm.

As for the woofer, it is a low pass network so it makes sense if you measure about 7ohm which is probably resistance of the driver voice coil and the inductor.


Hi Guys,sorry for the late reply, my wife doesn't see my urgency about this,so im going with the flow!!,please bear with me.
So,i now sat here with one of the crossovers multi meter in hand and i will try answer al the questions
DC offset,taken at speaker cable end ,speaker removed
repaired amp mid/high left chan 0.000 right chan 0.001v {DC]
bass amp                     Left chan 0.006 Right chan 0.002 {DC]
on xo IP pos and IP neg have good solder joint{tested good}
resistance between IP POS And IP neg   -OL ,
Multi meter checks out good,i have two of them ,both reading close enough.
No visabe shorts on board or speaker terminal board
mid speaker cable going to driver,dead short.no driver connected
tweeter spaker cable going to driver ,dead short, no driver connected
bass cable going to driver. OL,   no driver connected
ALL 3 OF THESE TESTS WERE ALSO TAKEN FOR XO PCB END
internal speaker cable from input to xo,No shorts

Test for shorts done at speaker cable imput {4 terminals}for bi-wire and Bi amp,inputs at the back of speaker boxes,   NO shorts 
to test resistors do i need to remove one side from pcb xo,ive just tested one [5w2R2J] and it showing a short,still fully soldered ??.
on the comments about capacitor placing on top of resistors etc,i do see what you mean,it was my first time replacing the caps on an xo and the leads where short,but i can remove the all and try again,its just that these caps are a lot bigger than was what installed at factory,sorry for my rubbish placement.
so all in ,i see no physical short, wires touching, all soldering has been tested for contin.all good,no dry joints,no cable fraying,no cable rub throughs,from vibration,it seem to me listening to you guys i maybe need to remove all components and test them,would i be wright in that assumption 
,hope the wife doesnt mind me spending the next day or two doing this !!!,shes wanting a  garage built !!!.
Anyway guys,i hope ive answered all you questions,i sure do appreciated you all taking the time out to help me with this,.
Kind regards
Mark
PS,im still open to tri-amping if you think it might be a better solution ,in the end,fiesta,but would be great to find the problem first,Anyway
cheers guys
resistance between IP POS And IP neg  -OL
Okay, so that checks out... as long as when you put the whole lot back together you still get open circuit at DC across the HF/Mid posts all is as it should be and there's no DC load on your amp.

mid speaker cable going to driver,dead short.no driver connected
Actually looking again you should be getting around 2.2Ω across the Mid driver outputs. I'd look at the photos again, but they're not there any more. The area to inspect is the bottom half of the board when orientated like the schematic.
Okay, its been a rough couple days. I just got done drawing the schematic and looking at the words and measurements again. Just one note, do all measurement in the resistance mode, not in the diode mode as the diode mode really doesn't tell us anything. You don't need to disconnect one end of the resistors to check each resistor. So with no drivers attached, between the midrange terminals on the crossover board, you get close to 2.2 ohms, but it's not a dead short correct? For the tweeter terminals, you get close to 0 ohms, but not a dead short correct? At the mid/hi amp inputs, you get essentially OL after holding the probes on the terminals for few seconds, correct? What happens when you reverse the leads on the mid/hi amp terminals? Does the meter go to (-) xxx.x ohms then to (+) xxx.x ohms before going to OL? This is a dumb question, but just so I understand everything. When you use only one amplifier to power the speaker in the full range (bi-WIRE) mode, the (+) terminals of the inputs are connected together. And separately, the (-) terminals of the inputs are connected together, correct? The photos are gone, but don't think we need it again. From your DC voltage measurements at the amps, they look healthy. Please answer all of my questions in order, so we don't get confused, thanks.
Which resistor, if it's STILL showing a dead short or 0.00 ohms? This is the one mentioned in post 6-16 @ 12:59am
CORRECTION, sorry. The last test on the mid/hi amp terminals on the crossover board; when I asked to to test and then reverse the meter leads and test again. The multimeter should be set on the "diode" test mode, not ohms. The readout should display some voltage then go to OL. When you immediately reverse the leads, the display should read -x.xxx volts, then change to +x.xxx volts before going to OL. Hope your meter has a diode test function, but it's not absolutely necessary. This test just shows that the caps are charging in both directions.
I still can't explain why the system worked with one amp when the mid/high amp was in repair... This is troubling.