Question for Jon

[xumbug]

Hi Jon,

I was just wondering if you had and further plans for the HiVi m12 drive unit in a 3 way. What are your opinions on this drive and would it make a good bass unit in a three way. What mid would you use, may be the m8a?
Your input will be much appreciated

Regards.

[JonMarsh]

Its funny you mention the M12a, because I was looking at it again for a possible project where higher sensitivity is desired. I wish they'd made an 8 ohm version you could parallel, and end up with a 4 ohm load, instead of using two in series if you need the Sd. The big question mark is the distortion behavior- if the motor design is of comparable quality to the M8a, then it's fairly good, but not state of the art by any means.

The M8a is not a midrange, really- I've used it in two way designs with a crossover point of 1250 Hz, works OK that way, but a driver like a Seas ER18RNX would make a better midrange. I like the inverted dome cone structure; it stays pistonic until all heck breaks loose, but even the HiVi Ma sees it's first cone mode at 4 kHz, which is lower than I'd care to see for a midrange unit. If I were to recommend a moderate price mid, it would be the RS100-4. You've already got a 4 ohm woofer, so you have to have an amplifier happy with low Z loads anyway. The RS100-4 has pretty reasonable response and distortion, especially for the money.









Problem could be sensitivity match, depending on the insertion loss in the woofer crossover, and the bandpass gain in the midrange- and the chosen crossover frequencies.

[xumbug]

Thank you Jon,

Would the m6a be better, maybe two of the in a mtm arrangment and then the m12 as a woofer. Sensitivity would be up in the mid section and also 4ohm load to match the woofer. Do you think this will work.

Regards.

[JonMarsh]

Yeah, sensitivity would be no problem, but I think the M5a would be easier to work with- the first breakup mode of the M6a is at 3 kHz. Check the Swans site for data- PE and Madisound are a bit derelict publishing complete specs.



Depending on your baffle step and voicing choices, you need a midrange roughly 87-89 dB/2.83 VRMS to work with the M12a. Also, depending on the inductor DCR and your crossover configuration, you may have another dB more midrange level relative to the woofer.

I have frequently almost bought an M12 to test just to see, but it's a bit more than an impulse purchase, and I used up most of my spare driver funds for this year already. :W Now I need to worry about crossover parts.

[HareBrained]

Zaph recently tested the M5a and that data may not agree with your recommendation.

The TB W5-1685 and SB15 are good performers. 150-4 is better than the M5a. Value of the PE buyout Peerless HDS 5.25" (830873) is unmatched. And as long as you're above 300Hz, the MCM 55-3870 has fairly low distortion.

[Evil Twin]

Saying it would work and whether the M5a is more extended than the M6a is not the same as a first recommendation- in fact, the RS100 would be the better choice.

It is a fact of life that many people buy and use the HiVi M series because of how they look, not how they measure. That is their choice.... there may be availability issues, too.

I find the RS150 less suited as a true midrange in a three way system- the larger RS series require steeper crossovers and trap filters, adding substantially to the cost and complexity of the crossover, and eliminating some potentially useful topologies. As a two way in a small MT, or MTM- it has its place. But energy storage becomes a problem from 1800 Hz up, like the RS180, and it responds best with crossovers under 2kHz. The RS100 can be used much higher with good results, and relatively slow roll off rates. I believe that is why Jonmarsh brought it up, to nudge Xumbug towards considering it.

OTOH, your suggestions and mine may be pointless because Xumbug is in South Africa, and his choice of drivers may be quite limited.

[ChrisGraeff]

Jon based on your prior usage of Hi-vi drivers would you presume the M12 to be usable up to ~700hz? Usable being defined (by me) as not requiring a quasi-elliptic LP XO? Specifically, I'm thinking about something 3rd order or 4th LR.

Probably should add I'd like to mate it with the TB 75-1558se.

[JonMarsh]

A few factors come into play for that question- my off the cuff answer would suggest 300 Hz as a more likely desirable limit. Though the major breakup mode appears to be centered at 3 kHz, there are substantial artifacts in the impedance curve below that. The next question is the audibility of that roughly 8 db peaking 3 kHz mode even when 12 to 18 dB down in the mix- is that sufficient? So if you trap it, and use an LR3, to get 2 kHz 18 dB down 700 Hz for -3 dB is doable. OTOH, my own experience is that I like to see the cone break up modes 30-40 db down- which is more like 300 Hz.

Another point I'm not familiar with on the M12 is resonance amplification of distortion products- I'd expect 6-10 dB of resonance amplification of the 3rd harmonic distortion from 1 kHz, which depending on the base line level might be fairly audible, as it is on many Seas Excel drivers in what otherwise looks like a reasonable range.

Last, the M12A motor does not appear to be using inductance control or copper in the gap- this argues for the likelihood of distortion increases with increasing frequency in the midrange- look at the inductance slope of the impedance.



If someone came to me tomorrow and said, "you've got to build a competitive DIY speaker with this driver and take it to a DIY shoot out and win, for minimum dollars", I'd go with either the RS150-4 and a low crossover point tweeter like the 27TBFCG or even using a sensitive 6-1/2" with extended top end like the ER18RNX. OTOH, for what the TB dome goes for, I'd consider spending a few bucks more to use something like the 6ND-430 as a midrange- very low 3rd harmonic from 200 Hz to 4 kHz for the most part. Of course, I'm speculating a bit about the M12a midrange distortion- not having measured it- call it informed speculation.

[bbe22]

To Evil Twin - our choice of drivers in South Africa is the same as everywhere else - it just costs a lot to ship them in - add 50% to double the US price in planning a project !

[HareBrained]

Nice write-up on your thought processes, Jon. Thanks. That's the sort of insights I look for while hanging around this forum. Based on my limited knowledge, your logic seems unimpeachable.

xumbug, what are your goals and constraints? Is there any line of drivers that are easier/cheaper to come by?

[xumbug]

Hi Every one,

My choice in drive units from Hi Vi is unlimited as a good friend works for them. Shipping to South Africa is the only real expence, but as the drivers are such good value direct, it makes it well worth it. To give you some idea, I can get the D8.8+ and D6.8 landed in South Africa for $90 and $65 respectively including shipping and taxes. The M12 by the way is used by Klipsch in one of their active subs the M8a is used in some of their other models (re worked cone) I know Klipsch are not ultra hi end, but good value non the less. Those drive units have got to sound better with a desent x over. I just think it would be interesting to see what people can come up with using these drivers.
Jon pretend you are going to a Diy with the intension of winning. Could be a fun 3 way project for people to build

bbe22, Where do you stay in SA. I Live South of Johannesburg in Henley on Klip