I was prompted to report on a little budget build I just completed. I had several goals:
1: Gain further practical real vs. model experience with PSD Lite and HOLM
2: Show how well you can do with cheap best-in-class components keeping a simple (cheap) crossover
3: Prove why I build more expensive speakers
I believe I succeeded in all three.
First driver selected was the SilverFLute W19RC36-8. At under $40, this is a wool-paper cone and is much better than the price suggests. No, it is not a Iluminator beater. It has terrible break-up issues between 4 and 5K so top prevent a very complex (expensive) crossover, it should be used with a very low crossover.It has pretty smooth response and will go quite deep for a 6 inch driver. I set 1.5K, 3rd acoustic as a max. Ouch. Find a 1.5K tweeter cheap? Again, picking best in class, not the best I can find, I selected the Dayton DC28F-8. 1/18 soft dome. Highly respected for a $20 tweeter and is known to not be too terrible crossed over low. Things like the SB29 or a Morel were way out of the budget.
Box was done by odougbo. It is one of his curved sid-port jobs he has posted several times. I don't know how he can do this so well without measurement, but the alignment is dead on. F3 is in the low 60's measured. Like I said, the SF is a budget class leader.
The crossover I ended up with is a second order electrical. Not quite symmetrical. Woofer is a 3.3mH and 12uF. Q is about 1. Tweeter 15uF, .75mH with a 1 & 8 Ohm L Pad. Q came out about .78. I used the Jensen P-core for the big 3.3. Caps are standard Dayton poly. For a strict budget build, I see no reason not to use electro's. I happen to have the films. The films raised the price by over $20. Pick your budget. I doubt you could hear the difference in this application.
One meter on axis, MLS measurements are within a couple dB 100 through 15K. The cabinet does not have the large radius edges I prefer so it sounded a tad bright. I added felt doughnuts around the tweeters to tame them a tad.
Sound.
Well, for me they are a bit disappointing. BUT these are built for under $200 the pair. For the budget, they are actually pretty darn good. The produce REAL bass into the low 50's. The response is surprisingly flat. Imaging is very good. The distortion is just too high for me. Yes Virginia, there is a difference between a $20 tweeter and a $100 one.
Red is the tuned speaker before adding the felt ring. I did not measure after. I did not do any polar measurements. Blue was a test trace I happen to have saved in this jpg. I was not planning to post this, so excuse the graphic.
Conclusion:
Through the crossover region, HOLM is a super tool and it correlates well with listening. I think it correlates better than fast-chirp on TrueRTA. It matches ARTA, but I prefer it now. Of course, the bass response is best done outside with noise, not MLS. Not going to bother, sorry. I use my WooferTester to get the impedance plots.
Using PSD, one can very accurately model repeatable results that correlate with measurements. Great little tool. I still rely on ISD for my box design.
The best budget drivers are pretty good. Only that. No amount of fancy crossover or tuning can overcome the lack of copper pole shunts and basic motor shortcomings. For $200, not a bad build. I would do a conventional rear-ported MDF box with 3/4 inch roundovers on all 12 corners. I may still do that. It may still be a tad bright for some if played stand-alone. With a sub, the balance is not bad. You can always pad the tweeter some, but I found that messed up the crossover region too much. So that means adding a LR parallel filter. More money which puts it outside the project scope.
What else could be done? I have another woofer I got by mistake. I may try doping the cone to see if the breakup can be tamed enough to slide the crossover up over 2K. That would greatly reduce the tweeter distortion and allow a cheaper woofer coil. Maybe I'll play with that while my test blocks for how to hide MDF seams cure.
They will stay in my garage playing the FM until I cycle something else through there. I have not run out of nieces and nephews to overload with prototype speakers yet so they will find a home.
1: Gain further practical real vs. model experience with PSD Lite and HOLM
2: Show how well you can do with cheap best-in-class components keeping a simple (cheap) crossover
3: Prove why I build more expensive speakers
I believe I succeeded in all three.
First driver selected was the SilverFLute W19RC36-8. At under $40, this is a wool-paper cone and is much better than the price suggests. No, it is not a Iluminator beater. It has terrible break-up issues between 4 and 5K so top prevent a very complex (expensive) crossover, it should be used with a very low crossover.It has pretty smooth response and will go quite deep for a 6 inch driver. I set 1.5K, 3rd acoustic as a max. Ouch. Find a 1.5K tweeter cheap? Again, picking best in class, not the best I can find, I selected the Dayton DC28F-8. 1/18 soft dome. Highly respected for a $20 tweeter and is known to not be too terrible crossed over low. Things like the SB29 or a Morel were way out of the budget.
Box was done by odougbo. It is one of his curved sid-port jobs he has posted several times. I don't know how he can do this so well without measurement, but the alignment is dead on. F3 is in the low 60's measured. Like I said, the SF is a budget class leader.
The crossover I ended up with is a second order electrical. Not quite symmetrical. Woofer is a 3.3mH and 12uF. Q is about 1. Tweeter 15uF, .75mH with a 1 & 8 Ohm L Pad. Q came out about .78. I used the Jensen P-core for the big 3.3. Caps are standard Dayton poly. For a strict budget build, I see no reason not to use electro's. I happen to have the films. The films raised the price by over $20. Pick your budget. I doubt you could hear the difference in this application.
One meter on axis, MLS measurements are within a couple dB 100 through 15K. The cabinet does not have the large radius edges I prefer so it sounded a tad bright. I added felt doughnuts around the tweeters to tame them a tad.
Sound.
Well, for me they are a bit disappointing. BUT these are built for under $200 the pair. For the budget, they are actually pretty darn good. The produce REAL bass into the low 50's. The response is surprisingly flat. Imaging is very good. The distortion is just too high for me. Yes Virginia, there is a difference between a $20 tweeter and a $100 one.
Red is the tuned speaker before adding the felt ring. I did not measure after. I did not do any polar measurements. Blue was a test trace I happen to have saved in this jpg. I was not planning to post this, so excuse the graphic.
Conclusion:
Through the crossover region, HOLM is a super tool and it correlates well with listening. I think it correlates better than fast-chirp on TrueRTA. It matches ARTA, but I prefer it now. Of course, the bass response is best done outside with noise, not MLS. Not going to bother, sorry. I use my WooferTester to get the impedance plots.
Using PSD, one can very accurately model repeatable results that correlate with measurements. Great little tool. I still rely on ISD for my box design.
The best budget drivers are pretty good. Only that. No amount of fancy crossover or tuning can overcome the lack of copper pole shunts and basic motor shortcomings. For $200, not a bad build. I would do a conventional rear-ported MDF box with 3/4 inch roundovers on all 12 corners. I may still do that. It may still be a tad bright for some if played stand-alone. With a sub, the balance is not bad. You can always pad the tweeter some, but I found that messed up the crossover region too much. So that means adding a LR parallel filter. More money which puts it outside the project scope.
What else could be done? I have another woofer I got by mistake. I may try doping the cone to see if the breakup can be tamed enough to slide the crossover up over 2K. That would greatly reduce the tweeter distortion and allow a cheaper woofer coil. Maybe I'll play with that while my test blocks for how to hide MDF seams cure.
They will stay in my garage playing the FM until I cycle something else through there. I have not run out of nieces and nephews to overload with prototype speakers yet so they will find a home.
I did listen to these for about 15miuntes; thought TVR had a bit of a break-through - the woofer sounded great with the larger coil, and TVR also mentioned light on the damping material.
Please remember they are less than $30, and sound like drivers 2 - 3 times the price.
Links: The Madisound Speaker Store
The Madisound Speaker Store
The box are very compact and I do think the side flare ports are enhancing the bass; we'd have to make up some other boxes to compare.
The only drawback I see is that they don't work well at low volume, need to play them at a moderate level. This is less of a problem in the mtm boxes.
Please remember they are less than $30, and sound like drivers 2 - 3 times the price.
Links: The Madisound Speaker Store
The Madisound Speaker Store
The box are very compact and I do think the side flare ports are enhancing the bass; we'd have to make up some other boxes to compare.
The only drawback I see is that they don't work well at low volume, need to play them at a moderate level. This is less of a problem in the mtm boxes.
Hi,
If you want a good cheap wide range speaker look no further than :
https://sites.google.com/site/undefinition/diy/amiga
Smaller this is pretty good at the price :
X-LS Classic kit
rgds, sreten.
Basic box design is so simple there is no need to use conjecture.
If you want a good cheap wide range speaker look no further than :
https://sites.google.com/site/undefinition/diy/amiga
Smaller this is pretty good at the price :
X-LS Classic kit
rgds, sreten.
Basic box design is so simple there is no need to use conjecture.
Last edited:
Been to the GR site a few times, their pricing seems to be good.
Impressive write up on the X-LS Classic, all that for a mere $109, seems to good to be true.
Wish they had more reviews, e.g. 53 comments on this PE page:
Dayton Audio BR-1 6-1/2" 2-Way Bookshelf Monitor Speaker Kit 300-640
TVR, what about a light spray vanish coat on the SF's?
That ploy is used by a lot of manufactures.
Impressive write up on the X-LS Classic, all that for a mere $109, seems to good to be true.
Wish they had more reviews, e.g. 53 comments on this PE page:
Dayton Audio BR-1 6-1/2" 2-Way Bookshelf Monitor Speaker Kit 300-640
TVR, what about a light spray vanish coat on the SF's?
That ploy is used by a lot of manufactures.
Trying to make an elliptical filter hit dead on a resonance peak would be quite a trick. It would have to be tuned to each speaker. I think the Q of the notch is much higher than the Q of the breakup and I have no experience with the phase. I have never explored this with a passive filter.
I played with a wide variety of Q's. PSD lets you do that. I have no doubt higher-steeper filters and a notch would reduce the tweeter distortion. Making the very best crossover possible with these two drivers was not the goal. Simple and cheap was. We already learned this is a great cheap woofer IF you deal with the notch. The Dayton has a pretty solid rep for $20. I have used several other $20 tweeters in the past. The Dayton is way up there in the pack and the only one I know anything about that can be at all civil at 1500Hz.
I would love to see a competition here on under $200 (US)/pair total cost on a mid-size monitor with 60 Hz f3 and reasonable power handling. A speaker for the masses if you will. Could we beat Warfdales for the price? I bet a lot of first time builders would love to see it. Take this as a first shot. There is a lot of experience in this forum and I have no doubt I could be bested. Don't be surprised If I don't follow up as it is just not what interests me right now. Here is an idea: Dope the cone to lower the breakup up Q allowing a higher crossover. I was unable to match the responce with only one resistor in the tweeter pad. Can someone do better? ( Buck-fifty savings) Do the poly caps matter? ( a ten spot to be saved). FG rather than wool stuffing? Anther few bucks. That almost pays for $30 tweeters.
Still, this was a simple build using commonly available best-in-class (really cheap) drivers and demonstrated learning how well PSD matches measurements. I already concluded that for my use, I need to spend the bucks on drivers about two steps up. Seas tweets at about $50 and woofers in the 60 or so range. Crossovers need about twice the number of parts.
When I get to my lab PC, I'll grab the four measurement files and throw them up.
Hi,
Elliptical is not really that difficult approached properly,
see : Zaph|Audio also see his tests of the 4 ohm SF.
rgds, sreten.
Last edited:
Quite familiar with Zaph's data. A most valuable place to start. Of course, one's own measurements and experience has to be added.
An elliptic is anything with a Q of much over 1.2 or so. Higher then Chebyshev. They are not really much use in low order filters and not common where linearity is important. Adding ripple to the pass band is not usually a help. If I remember correctly, Dick Small did a paper on them. I am not aware of any published successful designs using passive ones, and for sure, not second order. Note that the filter Q's are 1 and .78, so they already have some ripple. Play with the model if you wish, who knows what may be stumbled onto.
An elliptic is anything with a Q of much over 1.2 or so. Higher then Chebyshev. They are not really much use in low order filters and not common where linearity is important. Adding ripple to the pass band is not usually a help. If I remember correctly, Dick Small did a paper on them. I am not aware of any published successful designs using passive ones, and for sure, not second order. Note that the filter Q's are 1 and .78, so they already have some ripple. Play with the model if you wish, who knows what may be stumbled onto.
Amiga: Notice his comments on the VIfa tweeter. Truth be had there. 2K second order electrical is still low for the Vifa's. 3K is more like it. ( My surrounds use the XT25 @ 3K 2nd order with RS150's. Still not quite high enough but I get away with it as surrounds)
Dayton kit: I don't know the woofer so I can't comment. If the crossover was below 2K, I would be skeptical.
Dayton kit: I don't know the woofer so I can't comment. If the crossover was below 2K, I would be skeptical.
Hi,
No they not. They are defined by the fact the response in
stopband does not monotonically decrease it can rise.
The filter in Zaphs design is 1st order low pass followed
by a notch filter. Electrically it is Elliptical /Cauer, as the
response in the stopband rises above the notch.
Electrical transfer functions of Zaph's crossover
Note the BSC and bump in the roll off around 15KHz.
It is used in Zaphs design to approximate 4th order
L/R acoustic around the x/o point and then null the
drivers peak. The acoustic result is quasi 4th L/R*,
away from the x/o point the roll off is not 4th L/R.
Zaphs design is a successful use of of 3rd order
Elliptical filter, and 3rd order is the minimum
order for an Elliptical filter, so quite frankly
you've no idea what you are talking about.
rgds, sreten.
* Slightly assymetrical to control vertical directivity.
It is pretty effective, raw drivers in the box :
Adding the crossover (ignore sub 200Hz both graphs) :
Impressive stuff with just 3 components on a difficult to use bass/mid.
Last edited:
it wouldn't matter s long as the stopband attenuation level is exceeded any small ripples aren't a big problem.
lilun
Cool, thanks. Need the acoustic offset and baffle dimensions to play (or you could upload your PSD file).
Hi,
Whether it matters or not has nothing to with the definition.
Used properly of course it doesn't matter, proper use.
rgds, sreten.
Last edited:
OK. lets bet a bit more precise even though it is totally irrelevant to a budget build.
The definition of an elliptical filter is that it has equal ripple in both stop and pass. WIKI has very good discussion if it. You can build one by cascading a Chebychev type one and type two filter, which means you have to have at least fourth order. The Q will be greater than one.
If you build one with caps and coils, you will not get the infinite notch in the stop band. You only see that with "perfect" active crossovers.
The Elliptical has the fastest initial rolloff of any analog filter. But it is still the same slope as all others in the same order. Only the knee is sharper. This does nothing to keep away from breakup modes two octaves higher. The steeper slope is what would help.
Now, back to a budget build. I would not build a baffle the way these were. The side ports preclude the large edge radius I favor and I have concluded offset tweeters are more trouble than they are worth on small baffles. I prefer regular polygon cabinets with rear ports and I like to cut the flange of the tweeters to get them closer to the woofer to improve horizontal off-axis. Things can be gained with irregular shapes, but they are not really relevant at this end of the scale.
Again, It would be great if someone wanted to run with a best under $200, F3 of 60 Hz, under 3/4 cubic foot build-off. I would be very disappointed if the talent on DIY could not better these. I probably could. This was an excersize with the SilverFute, a better than expected for the price cheap woofer. Don't expect more from it.
For every increase in dollar spent, does it allow it to be recovered elsewhere?
So, if you went third order and crept the crossover up to , maybe 1900, yes the tweeter would be happier, but where are you going to recover the cost of extra coils and caps?
Is there a cheaper tweeter that would work better? I seriously doubt it but maybe one for $ 2 bucks less could be found. This is a darn good driver for $20.
Can you find a poly cone that could be crossed higher giving more freedom in a tweeter? Maybe, but I find cheap poly cones sound very lifeless compared to cheap paper. The Silver Flute really has more solid bass than expected in a small-ish box with reasonable distortion. It has pretty darn smooth response in it's usable range. Can you beat it? Please do tell.
Could doping the cone help the breakup allowing a higher crossover at no cost?
Is changing to electro caps enough savings to buy better tweets, or to pay for a notch allowing higher crossover?
Can you save a buck by using one resistor in the pad?
Go forth all ye fine folks and do me better. In the mean time I have some bamboo cabinets to build and some MDF finishing techniques I am trying.
The definition of an elliptical filter is that it has equal ripple in both stop and pass. WIKI has very good discussion if it. You can build one by cascading a Chebychev type one and type two filter, which means you have to have at least fourth order. The Q will be greater than one.
If you build one with caps and coils, you will not get the infinite notch in the stop band. You only see that with "perfect" active crossovers.
The Elliptical has the fastest initial rolloff of any analog filter. But it is still the same slope as all others in the same order. Only the knee is sharper. This does nothing to keep away from breakup modes two octaves higher. The steeper slope is what would help.
Now, back to a budget build. I would not build a baffle the way these were. The side ports preclude the large edge radius I favor and I have concluded offset tweeters are more trouble than they are worth on small baffles. I prefer regular polygon cabinets with rear ports and I like to cut the flange of the tweeters to get them closer to the woofer to improve horizontal off-axis. Things can be gained with irregular shapes, but they are not really relevant at this end of the scale.
Again, It would be great if someone wanted to run with a best under $200, F3 of 60 Hz, under 3/4 cubic foot build-off. I would be very disappointed if the talent on DIY could not better these. I probably could. This was an excersize with the SilverFute, a better than expected for the price cheap woofer. Don't expect more from it.
For every increase in dollar spent, does it allow it to be recovered elsewhere?
So, if you went third order and crept the crossover up to , maybe 1900, yes the tweeter would be happier, but where are you going to recover the cost of extra coils and caps?
Is there a cheaper tweeter that would work better? I seriously doubt it but maybe one for $ 2 bucks less could be found. This is a darn good driver for $20.
Can you find a poly cone that could be crossed higher giving more freedom in a tweeter? Maybe, but I find cheap poly cones sound very lifeless compared to cheap paper. The Silver Flute really has more solid bass than expected in a small-ish box with reasonable distortion. It has pretty darn smooth response in it's usable range. Can you beat it? Please do tell.
Could doping the cone help the breakup allowing a higher crossover at no cost?
Is changing to electro caps enough savings to buy better tweets, or to pay for a notch allowing higher crossover?
Can you save a buck by using one resistor in the pad?
Go forth all ye fine folks and do me better. In the mean time I have some bamboo cabinets to build and some MDF finishing techniques I am trying.
Oh I don't know, I thought they sounded like the King's Trumpets when I heard them, wish I could have kept them for a few days.
What about the next Dayton up the chain? I'm thinking they have more of the $100 tweeter sound. They supply them with truncated rings.
Have two of everything here, woofers, boxes, tweeters, let me know if there are any suggestions for designs.
What about the next Dayton up the chain? I'm thinking they have more of the $100 tweeter sound. They supply them with truncated rings.
Have two of everything here, woofers, boxes, tweeters, let me know if there are any suggestions for designs.
Attachments
OK, now you are going to make me put them in a box with my style of large radius baffle and see if I can tweak the top end a tad more. My budget won't hold up to a set of the Santori drivers just yet, so I'll play here a little. Do note, we have two designs now using the Santori's. Both by very experienced designers. (see the vendor section). It took over a year for them. It is going to take me quite a bit if time just to absorb all the info in the thread. Very good reading. This build was a few days. Budget in both time and cost.
I will mount the drivers on a baffle intended to manage diffraction better.
I will measure acoustic offset. I use an old Old Colony Lab "pulser" for this but it can be done with a PC and Zelscope.
I will provide new 1M plots in a quieter environment with consistent gate times.
I normally would not splice in the LF responce. It is too room sensitive anyway. Maybe I will try. SoundEasy allows doing a Hilbert Transform to generate the phase of a spliced plot. I understand it is a comfort factor for beginners to see something on the bottom end that looks normal rather than the artifacts of a gated MLS plot. It does not matter to crossover design.
I will set a measurement reference level that can be reproduced instead of "about a WATT"
Doping the cone to see if I can get a higher crossover by controlling the breakup is not a beginners technique but it is cheap. I will only include it if I can get stupid-lucky first time out. ( I have one extra woofer) You can trash a lot of speakers learning how, so budget it is not. Mess up only one pair, and you could have bought a much better drivers in the first place! ( like the Seas woven poly cones with no breakup problems, copper pole shunts etc. )
I will do a BOM to set a cost (sale prices and what is in the junk box are cheating) to enforce "do better cheaper"
I will assume any complex cabinet shapes, but if that means CNC work, it has to be included in the price. Anything I can do with a table saw can be done with a hand saw. It is just skill and time. If you want to build a dodecahedran, be my guest. I will build a PB box with a cardboard tube port. for impedance measurements.
I will finish off my cheapskate measurement lab guide. I got distracted building a new bathroom this summer. As LIMP can't do PayPal, I never bothered to buy it and just use my WooferTester for impedance measurements. It is a great tool, but not ultra cheap so it did not follow under cheapskate guide. LIMP does. SoundEasy does everything, but it is not beginners cheap either. Look for the post on the PSD-Lite thread. in a few days.
All the software costs about $25 and building a mic as Sigfried shows is maybe another $10. Maybe I'll even do the postal money order to get the LIMP license. Zelscope is about $12 bucks I think. There is NO EXCUSE not to measure! and simulation software is so cheap, there is no excuse not to model.
Time for coffee and get to working. I will not spend any more money for the next round than the first round cost. So don't expect exotic high order crossovers. IF using electro caps saves enough to go third order or add a notch, that is fair game. Hard line on the cost.
I just found out how I keep losing text on the screen. I am bumping the touch pad with the side of my thumb on my new laptop. It makes the curser bounce around. I use a real mouse, so I am going to cover it up.
I will mount the drivers on a baffle intended to manage diffraction better.
I will measure acoustic offset. I use an old Old Colony Lab "pulser" for this but it can be done with a PC and Zelscope.
I will provide new 1M plots in a quieter environment with consistent gate times.
I normally would not splice in the LF responce. It is too room sensitive anyway. Maybe I will try. SoundEasy allows doing a Hilbert Transform to generate the phase of a spliced plot. I understand it is a comfort factor for beginners to see something on the bottom end that looks normal rather than the artifacts of a gated MLS plot. It does not matter to crossover design.
I will set a measurement reference level that can be reproduced instead of "about a WATT"
Doping the cone to see if I can get a higher crossover by controlling the breakup is not a beginners technique but it is cheap. I will only include it if I can get stupid-lucky first time out. ( I have one extra woofer) You can trash a lot of speakers learning how, so budget it is not. Mess up only one pair, and you could have bought a much better drivers in the first place! ( like the Seas woven poly cones with no breakup problems, copper pole shunts etc. )
I will do a BOM to set a cost (sale prices and what is in the junk box are cheating) to enforce "do better cheaper"
I will assume any complex cabinet shapes, but if that means CNC work, it has to be included in the price. Anything I can do with a table saw can be done with a hand saw. It is just skill and time. If you want to build a dodecahedran, be my guest. I will build a PB box with a cardboard tube port. for impedance measurements.
I will finish off my cheapskate measurement lab guide. I got distracted building a new bathroom this summer. As LIMP can't do PayPal, I never bothered to buy it and just use my WooferTester for impedance measurements. It is a great tool, but not ultra cheap so it did not follow under cheapskate guide. LIMP does. SoundEasy does everything, but it is not beginners cheap either. Look for the post on the PSD-Lite thread. in a few days.
All the software costs about $25 and building a mic as Sigfried shows is maybe another $10. Maybe I'll even do the postal money order to get the LIMP license. Zelscope is about $12 bucks I think. There is NO EXCUSE not to measure! and simulation software is so cheap, there is no excuse not to model.
Time for coffee and get to working. I will not spend any more money for the next round than the first round cost. So don't expect exotic high order crossovers. IF using electro caps saves enough to go third order or add a notch, that is fair game. Hard line on the cost.
I just found out how I keep losing text on the screen. I am bumping the touch pad with the side of my thumb on my new laptop. It makes the curser bounce around. I use a real mouse, so I am going to cover it up.
Next Dayton is the same tweeter for $1 more to get a truncated frame. Next REAL step is $54 each vs. $19.45. Clearly out of scope for this game. Thread comments suggest it is a good tweeter. I am a Seas fan at that price.
To tame the top end just a little, Radius the top edge 3/4 inch and add a felt doughnut around the tweeter. You can't do much to the sides. In your room, move all the other speakers away so they don't give as many reflections. Rule of thumb is 1 meter in all directions. Of course, all our thumbs are different.
Maybe change the L-Pad to drop the tweeter a dB. I thought it started to suppress the mids a bit which is why I left it hot. Adding a counour filter is upping the cost. Not allowed unless it is recovered elsewhere.
Tweeter offset/horn loading. Nothing helps a tweeter like efficiency. Carving a horn is not easy. 3/4 roundover if it can line up with the surround having no gap may be an approach. Let me look at that. Of course, it will take some crossover tweaks that could imply cost. After the conventional rear ported box, that would be my next trick. I wasn't going to let that one out, but whatever.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.