DIYSG HTM-12v1 Speaker Review

[richard12511]

never going to get smooth from a compression driver on a horn

JBL M2 and 705/708 seem to do it pretty well. Don't they use compression drivers?

 

[jtwrace]

JBL M2 and 705/708 seem to do it pretty well. Don't they use compression drivers?

Yes and all the other best measuring ones.

 

[ooheadsoo]

Yes and all the other best measuring ones.

Corrections to compression drivers can obviously be done, but driver to driver consistency of the amount being displayed here is going to wreak havoc on small DIY operations. Here's to hoping the Celestion CD in v2 has tighter tolerances. For these speakers and for their intended application, applying EQ/room correction should be standard procedure, anyway.

I thought the JBL M2 is only offered without passive crossovers, for the end user to have their choice of active crossovers.

 

[jtwrace]

The B&C DE250 is extremely consistent and one of the many reasons they’re popular whether diy or not.

Yes, M2 is active.

 

[beaRA]

I do not think so, others will know more. But even without direct entry of PEQ parameters, you could match the required EQ curve. Something like this might help:
https://www.musicdsp.org/en/latest/Filters/218-plotting-r-b-j-equalisers-in-excel.html

I don't think that will work. Even if you recreated the filter response in the curve editor, you still have to let Audyssey measure your in room response to try to match the target. My goal for anechoic correction would be to optimize the LW response and then allow the natural room response above the transition frequency.

 

[jhaider]

JBL M2 and 705/708 seem to do it pretty well. Don't they use compression drivers?

Yes and no. JBL’s patented phase plug seems to be beneficial. However, without the active EQ they’re not very smooth. EQ takes care of some resonances and the end result is smooth. Unlike most I've measured 705i without EQ in the chain, so I've seen the raw response.

 

[changer]

Can you tell us how smooth is defined? I would think it is defined as variance in SPL in relation to Q/bandwidth, but this is just a guess. There is probably a technical definition by some association which maybe even relates driver performance to psychoacoustics/ERB, but I do not know a paper.

 

[johnp98]

Random question about measurements and EQing this speaker.
My understanding is that below the Schroeder frequency (100-300hz for most rooms) the sound is modal and room dependent and thus prime for EQing out the peaks.
It also looks like gated measurements are good for accurately measuring the high frequencies.

So.... if we know based on the klippel measurements that this speaker has very good DI and off axis consistency then could one do gated on axis measurements for their own HTM-12 (as this would measure and account for their specific driver variations) and then EQ the highs based on the gated measurements and then EQ the lows based on the in room response?
Any idea the best way to measure the midrange?

Seems like it would not be too hard and then would give you a great result based on your own speaker and room (with the assurance based on the klippel that the off axis will follow your EQ based on the on axis).
Thoughts?

 

[ooheadsoo]

I like the "power hifi" moniker, but I thought this was designed first for home theater. As such, it should be sensitive, low distortion and have good speech intelligibility.

This specific iteration was intended for HT, but the SEOS portion was developed after years of Geddes espousing his work on the OS horn shape in the DIY community, and the DIY kit community finally taking a stab at making it available to more DIYers instead of sourcing only via Dr. Geddes's hand crafting. I remember not being able to afford any of Dr. Geddes's iterations, back in those days. In the development process, the organizers decided to go the SE route for various reasons, with smaller vertical spacing for a larger vertical lobe being primary. That it happens to be great for HT is, IMO, somewhat incidental.

The poor measurement seems to be a result of poor driver manufacturing tolerances and not intended by design. Active implementation, which is basically the same thing as applying EQ, does seem to be the best way to go, but that's probably true of any design at the end of the day.

 

[Absolute]

Yes and no. JBL’s patented phase plug seems to be beneficial. However, without the active EQ they’re not very smooth. EQ takes care of some resonances and the end result is smooth. Unlike most I've measured 705i without EQ in the chain, so I've seen the raw response.

How did it look? I have some measurements of the horisontal response in the M2 waveguide. Not sure how I'm supposed to measure that so I did it two ways - one where all the angles were measured with laser to the same distance from center driver and one where I just moved the mic in a flat line to the side. This is how it looks (EQ'd to flat on-axis for reference).


Same distance to center driver in all measurements, gated 3 ms;

Sliding the mic in a straight line to the side for each measurement

And an average of all those combined relative to a flat line;

 

[changer]

In the development process, the organizers decided to go the SE route for various reasons, with smaller vertical spacing for a larger vertical lobe being primary.

Pardon, what is the "SE" route?

 

[ooheadsoo]

Pardon, what is the "SE" route?

Super ellipsoid.

 

[Deleted member 12642]

I built these pluse some volt 10's into some of my own cabinet designs, more suited for live use. The htm's sit on poles on some tham15's and the volts are for monitors. They turned out great in terms of build but the sound quality on both is just super meh. They both work pretty well with hefty EQ but that means more rack gear to haul. I'm hoping to toss some DSP+amp combo board from parts express into the volts. I very much like them as monitors. The HTM12's will be sold for sure.

 

[ooheadsoo]

I built these pluse some volt 10's into some of my own cabinet designs, more suited for live use. The htm's sit on poles on some tham15's and the volts are for monitors. They turned out great in terms of build but the sound quality on both is just super meh. They both work pretty well with hefty EQ but that means more rack gear to haul. I'm hoping to toss some DSP+amp combo board from parts express into the volts. I very much like them as monitors. The HTM12's will be sold for sure.

How did you design the cabinets without the t/s parameters? The crossovers were designed for 2pi space, I somehow don't think that would translate well to what sounds like a pole mounted 4pi situation.

Fwiw, I like my HTM-12 v2.

 

[hardisj]

Good news! Matt is shipping me the parts to upgrade the loaner v1 to a v2 so I can test it. Stay tuned…

 

[johnp98]

Good news! Matt is shipping me the parts to upgrade the loaner v1 to a v2 so I can test it. Stay tuned…

That is fantastic! I am very excited! Feels like all the effort warrants a donation!

 

[Deleted member 12642]

How did you design the cabinets without the t/s parameters? The crossovers were designed for 2pi space, I somehow don't think that would translate well to what sounds like a pole mounted 4pi situation.

Fwiw, I like my HTM-12 v2.

I just kept the volume the same, they sound pretty much the same indoors as they do on poles. I don't really care for them though, they're just a pain to use when I have other powered option with DSP that just sound better. The volts are close to the best speaker monitoring I've ever had with EQ, but without it they kind of suck. Here's one volt and a tham. Don't have any pics of htm.

 

[ooheadsoo]

I just kept the volume the same, they sound pretty much the same indoors as they do on poles. I don't really care for them though, they're just a pain to use when I have other powered option with DSP that just sound better. The volts are close to the best speaker monitoring I've ever had with EQ, but without it they kind of suck. Here's one volt and a tham. Don't have any pics of htm.

View attachment 132196

From what I recall, the Volt measurements aren't great for FR linearity. However, they (and some of the other pro coax alternatives that are actually in stock) are so relatively inexpensive, I could be tempted to try an active implementation, just for fun.

Did you make HTM 12 v1 or v2? I can easily see choosing a different speaker for PA just from the weight of a wood cabinet of this size alone. I don't know if I would call my v2 "meh", though. I like them - but I use them at home and with DSP.

 

[Deleted member 12642]

From what I recall, the Volt measurements aren't great for FR linearity. However, they (and some of the other pro coax alternatives that are actually in stock) are so relatively inexpensive, I could be tempted to try an active implementation, just for fun.

Did you make HTM 12 v1 or v2? I can easily see choosing a different speaker for PA just from the weight of a wood cabinet of this size alone. I don't know if I would call my v2 "meh", though. I like them - but I use them at home and with DSP.

It's version 1.

They don't weigh anymore than most wood 12" two ways.

 

[peanuts]

the funny part is that over on avsforum they have been bragging about the great measurements of all these speakers for years, and they cant even make one that is remotely close to flat. LOL
i knew they were not to be trusted.