Hi everyone, sorry for the delay in getting more info to you, but i've had to doso much in the last few days, that I've onlybeen able toglance at the number of notifications in my inbox, let alone read them!. WFMOB, if I thought that there would be this much interest in the project I would have posted with a little more preparation!.
There are two reasonswhy I haven't posted the driver yet, one being the reliability of a new driver that is as yet untried in the field by us end (ab)users.
Second is price and availability.
As fortheposts I have made so far, I have never claimed that this driver is better than any other. I have only made comparisons between twodifferent horns loaded with twodifferent drivers. My conclusions based upon the results of preliminary testing are purelysubjective, highly biased, and are of course just my personal opinion!.
However, here we go, down to the bare essentials. This is going to be somewhat of a peroration, and I apologise for the length.
The plan was instigated after my son decided he wanted to put a small rig together for him and his friends. We built two 1850 horns and used Void V1000 drivers. I swore (hundreds of times!) then, that there had to be aeasier way of making a horn, Due to the complexity of build, and the almost impossible task of fitting the driver!. Never again!.
So this project has been ongoing for more than 18 months now, and after numerous complete and dismal failures, a good few thousand pounds, and much time, sweat and tears, we have come up with a result. (my son makes his own contribution of plans, ideas, and helps screw the things together.)
After sifting through hundreds of different driver parameters, modelling nearly all of them, and rejecting most immediately, I have to admit that my own personal prejudices made me reject this driver without even modelling it!. (six months down the drain!)
The objective was to build a horn that was reasonably small and transportable, at least 130db across the lower bass spectrum between 40 and 90 hz, easier to build than most of the current designs, and as cheap as possible.
The standard to be used for comparison was (of course?) the 1850. Match the output andgo lower. (didn't want much, did we?!).
Cost dictated that the 1850, Void V1000 etc. were out, not many of uscan afford that price bracket. So the next best thing looked like the PD186. Either that or something better, with a lower price tag. If I wanted to get the price down I was resigned to buying 8 at once direct, to get down to about £128 ex VAT.
So with the PD186 in mind, where to start?.
Using only the resources that us mere mortals can afford, Hornresp is an invaluable tool.
The 1850 / 186horn never models according to specs, on Hornresp. After version 17.XX the bass end did improve dramatically, but the assumption made in the model that the response of a driver is linear, does mean you have to approach results with caution, and extrapolate from the predicted response, against some measurements taken independently.
So with this in mind, the 1850 horn parameter modelforHornresp looks something like this:
Put your own values in whereyou disagree with mine.
VRC: 80/ 65 Half space 700w (74.83 v) into 8
LRC:20cm
VTC: 23000
ATC: 1150
The horn may be modelled using four (exp/con) sections or one hypex section, the single hypex is just the closestextrapolation from the four.
4 SECTION EXP
S1: 670
S2: 1200 con/exp: 69
S3: 1800 con/exp: 41
S4:3000 con/exp: 41
S5: 5160 con/exp: 31
HYPEX
S1: 670
S2: 5160 hyp : 180 T 0.6-0.8?
The above is purely a resemblance only, and I have been generous in the overall length of the horn,projected out from the bottom edge and used the floor as part of the horn.
You can see the shelf between about 65 and 90 hz which corresponds quite accurately with
available specs for the PD186. However, the sensitivity at 1 watt is a bit low at 103 db.
OK. We have a basis for comparison. Accurate? no! but if we maintain theassumptions used to gain this reference point, and then place other parameters to plot against it, we can infer from the reference point which way changesmay effect the response from the new plot. As I said,I was generous with the length, and I do not like the extreme angle the 1850 horn opens up at. (just my opinion, not a criticism, that horn is good). I wanted to see more horn fully enclosed within the cabinet. To get down to 40 hz, thehorn needed to be exponentialin order to lower the frequency, andreduce the mouth area to throat area ratio. So the S2 was reduced to 4280 andTset to 1 for exponential. Now theplot for a 186 looked a tiny bit better, but the change did not justify a eureka moment.
I had also come to the conclusion that the 40 hz target wasimpossible. None of thedrivers could take the excursion at about 60 hz when the horn was tuned too low.
Back to find another driver. 1850 and void looked much better, although a lot lumpier.I considered a straight outexponential at 750 or 800for S1 and a smaller mouth between 2400 and 3000, looked nice and flat butsensitivity would be way down. Also cone excursion looked a lot worse. 1850 horn would still be louder over most of the range.
So start going back over the discarded manure heap and see if I had missed something.
After many hours of comparisons looking for that magical combination of cms rms xmax rms power etc. I thought I had made a mistake in entering the parameters at first, and spent ages checking web site figures, myfigures, and still had the same result.
Parameters:
SD: 1130
BL: 22.11
cms: 1.0E-04
rms: 7.72
mmd: 177.4
LE: 1.81
RE: 5
Significant?.
The rest is boring design work shoe-horning that size of horn into as small a box as possible. Wrapping the horn around the rear chamber and using as much of both sides is the most efficient way I could find to do the job. However, the parameters for the horn remain mostly unchanged from the T1 1800 exponential. The end results after building the prototype were even better than the predicted plot. Against the Void, which tends to drop off at 80hz to 90hz, this driver is flat. At the lower end going down past 50hz the void really drops off, but this driver still keeps going down to 40hz. It does follow the predicted angle of slopefrom40 to50hz, but this levelsoff at 50 all the wayup to90hz.The 3db gap between the two is very noticable. It is not perfect, it is still less than what we wanted to achieve, but it does give usable, audible response down to 40hz, albeit 2db down from what we really wanted. (please note this is all for a single horn in half space) multiples of two or more will acheive the magic 40hz more effectively!, and if you take a point between half and quarter space plots you will get an idea of what the response would be in a club room. Now to the nitty gritty bit of cost. After finding this little gem I was rather let down by the fact that no one has got any. Oh !?%$ !!!. The range is being advertised by the dark side, but I have been informed by a third party that the 18s are the only ones that are currently in production, andthe darkoneshave taken 6 weeks, and still haven't ordered any yet!. (where have I heard that before?) So it's bite the bullet time and get a mortgage.Final price to be confirmed, but is in the region of £115. Delivery next week. I am planning on building 8 cabinets as soon as possible, for me!. More to follow, but if any of you might be interested in having some of these drivers, I would be prepared to launch COMMUNITY AUDIO SUPPLIES and let you share the low cost benefits. (see community spirit post) Iam building a 'pre production' version with internal bracing on the inside of top, rear and bottomexternal panels to the rear chamber, and inside the chamberas well. The access panel is to be made much shorter, and give just enough room to load the driver. The throat chamber and start of horn will also be re-configured, but parameters will not be changed, just make things a bit easier to make, with less colourful vibrations. The internal panels do have angles, thay are not square. But there areno angles above 45. (don't tell me you expected it was finished as in the pics!). Plans will be available when pre-pro cabinet has been completed and tested.Further testing of the driver which has now clocked up nearly 200 hours. We have been using the cheapest amps that are capable of delivering the power in an attemp to give as realistic a test as possiblethat will reflect the abuse thiswill undoubtedly recieve in the field. The Citronic ultima pro 3200 gives a very good showing. Wound right up it does perform extermely well. DO NOT USE ABOVE CLIP POINT. Don't know how long things would last if continually run withboth channelsdriven to the limit.A distinct sensitivity improvement is noticable comparing the two horns using the same levels, same output, just swapping the output lead from one to the other, and measuring using same mic at same position etc. Less power needed, more bang from the new horn.Numark dim 4s show well in bridge mode, as does the berry EP2500. Sound quality is not 100% when these amps are driven hard, but what do you expect?. In all, the driver is capable of living through the normal abuse, but as with everything else, pass the point of no return and everything dies. Xmax appears to be reached at 100v peak to peak,using 40 hzsine wave. Approximately 700wrms. However, whenplaying music, I have yet to see (feel) the cone touch thexmaxprobe when driven with much higher levels. (yes I have to put up with everything at excruciating levels including techno, d+b, or whatever else my sons' hard core acid head friendswant to listen to). Sorry i'm showing my age, but I prefer some Free, Robert Plant or Van Morrison to kick back with.All will be revealed next week after I have the drivershere. Do not start flooding the posts with 'this is a xxxxxx': queer the deal and we all lose out. Just to say thanks again for the interest shown, I will post more when things are finalised. In the meantime feel free to load up your favourite driver and see if it is a contender for speaker of the year!.Again apologies for the length.
