I'm thinking over building DIY versions of the Performance Series with biased networks. My biggest challenge is that the only drivers currently available are the 908 Ti and the 906 Ti.
Are the other drivers (904 Ti, tweeter, LE14-H3) likely to be available sometime in the future?
"Zobel is as zobel does"
They've been available in the past. I've seen them go in and out of stock a couple of times. Since they're both used in current production, I'm convinced we'll see more of them.
I'd love to see your biased network plan.
<--------------What about the tweeter and EOS Waveguide?
Out.
Thanks - I'll keep my eyes open and snag when they are in stock.
Just looked over the technical manual - didn't realize the waveguide was separate from the tweeter. I have the part number (335695-001). That could be a show stopper unless I can:
1) Buy from Harman as a part -or-
2) Find a way to make my own -or-
3) Find a suitable replacement
If the project is a go, most of the crossover components are standard or close to standard values. Biasing the caps should be straightforward. I may need to purchase stock networks to get the DCR's from the inductors if noone knows them.
The size of the biased network is one of the main reasons I'm looking to DIY.
"Zobel is as zobel does"
Here's a start at that biased network. DCR's are .5 Ohm until I can get measured values. I've retained the bypass caps from the original PT800 network. Depending on cap selection they may not be necessary.
Note: L1, L3, and L4 are iron core per the original schematic.
Any CircuitMaker 2000 jocks out there who can show me how to make an iron core inductor symbol?
EDIT: I've updated the schematic to show the DCR's and cleaned up a few errors that snuck in last night.
"Zobel is as zobel does"
" I may need to purchase stock networks to get the DCR's from the inductors if noone knows them."
Pm me if you would like to know what I measured on the stock inductors.
L4 is an aircore on the stock crossovers that I have.
Baised networks aside, I didn't see the DIY approach to the PT800's being very cost effective given what you can buy them for.
What can they be had for, besides the $1050 on the Harman web site?Originally Posted by mikebake
I think $1050 is a good deal, frankly.
I got mine from an authorized distributor for quite a bit less. I'm thinking you might find new ones for under $800.
There are also used ones now and then on ebay. When I added up the drivers, crossover, grill, and money to have cabs constructed, it looked like a wash to me. I will likely pursue the biased network sometime in the not too distant future. Someone posted recently that they guessed the Performance Series started out as a low end product, but possibly was priced higher when the sound exceeded expectation. I would guess not, as I would think the designer knew pretty early on what the drivers were going to be capable of. They are rather nice pieces, and fine additions to the league of artful JBL transducers.
I also agree as above that DIY could offer a higher quality cab, but perhaps not gaining anything sonically.
I agree that $1050 is an excellent value. Every Forum member should buy a pair and do their part to keep JBL in the quality consumer business.
A PT800s DIY project is not for everybody. I happen to like building my own speakers and DIY is always the first thing that comes to mind when I want new ones.
Costs are still pretty attractive:
$350 for drivers
$200 for Solen based biased non-bypassed networks (add $70 for Audio cap theta bypasses)
$300 for cabs and hardware including a nice veneer.
I still see $150 ~ $200 savings potential plus the additional benefits of the biased network. If using the stock network, savings increase to ~$350.
Back to tech issues...
I've been able to find near matches for three the inductors using the DCR's provided by sceptical1. As Giskard predicted, I was able to find these values using Solen parts.
L2 is a bit harder. It is 0.06 mH with a 0.4 Ohm DCR. My fix is to use a 20 gauge 0.1 mH Jantzen and unwind as needed. I estimate the resulting DCR will be about 0.1 Ohm. To compensate R3 can be increased to 2 Ohms. The voltage divider of R3 and R4 really won't be affected much either way, so the stock 1.8 Ohm value willl work just as well.
"Zobel is as zobel does"
I had been thinking about using splines, but that is the ticket
"Zobel is as zobel does"
One thing you might want to consider. The 800 has a lot of sharp edges due to it's shape. Now the shape is due to the fact that the 800 is for a modular HT design for on walls, in corners and on the 1400, really cool design EXCEPT that those EDGES really screwup the sound if you are going audiophile....I've been working with "removing" the edges and the sound is so much more clear, 3D,...far less haze
when we removed the standard connectors between the 1400 and 800 a LOT OF HAZE disappeared and more clear music came thru.
The next thing I did with the 800 and 1400 connected was to eliminate the space ( using materails) between the two cabs and try to make it into "one" Box... again more haze disappeared..about 1/3 as much as from removing the connectors, however more things happened also, the lower mid-bass ( 100 to 400-500) expanded and filled out.. More black space btween the notes and more 3D notes. placement of instructments much better...
then I started working to "round" out the edges ( sides and top)...more changes..better but more subtle..yet powerful changes...
really makes me wonder what would happen if i were to make one box, with time alignment, completely rounded corners with the best baffle material and coverings, CC crossovers... so far I've not seen any limits to these speakers...if only I knew more about this so I could design this myself...
Giskard.....you give me way too much credit if you think I know enough of what you're talking about there or how to apply it.....
I know enough to be real stupid...
I'm not an electronic person, I'm not an audio engineer and REALLY, if you see me with a cutting tool in my hand nothing good can come of it.....
Uncle paul:
I have some questions on the crossover schematic. I'm not real bright on this Kind of stuff so you'll have to excuse me here.
I've been wanting to redo the crossovers in CC so this really is important to me. one thing about what you've listed I'm not understanding.. I've always thought that for CC you would
1. double the value of the caps and
2. double the number of caps
In your schematic ( for the low crossover) you show one cap ( c1a) at 100uf and the sercond cap ( c1b) at 130. The original schematic for the 800 shows one cap at 56uf and I thought that would translate to two caps at 112.
In the mid and high schematics you list the same thing happens..
so I quess my quess is why the different values for the caps and not just two with the same doubled values.
The answer is likely which specific values of caps are available from the parts' suppliers .so I quess my quess is why the different values for the caps and not just two with the same doubled values.
(i) Net capacitance ( of series caps ) = the reciprocal sum of all the caps in the series .
(ii) An example ; 100 uf & 130 uf in series = 56.52 uF ( effective capacitance )
> 100 ( its' reciprocal = .010000000 )
> 130 ( its' reciprocal = .007692308 )
> Sum = ( .017692308 )
> Reciprocal of ( .017692308 ) = 56.52173913
- Obviously; 2 , 112uF caps in series will give a capacitance of 56 uF .
(iii) Another example ;
> 112 ( its' reciprocal = .008928571 )
> 3 of these identical caps in series = .008928571 x 3 , giving a new total of .026785714
> Reciprocal of ( .026785714 ) = 37.3333333
> So, 3 , 112uF caps in series will give a net capacitance of 37.33 uF .
Exactly.
Everyone should have their biased capacitor matrix at the ready.
There are currently 1 users browsing this thread. (0 members and 1 guests)
Posting Permissions
Reply With Quote