Last edited:
Home-Made IEMs
- Thread starter Bilavideo
- Start date
High pass the 2389D with a capacitor, but then I think you want the full range on up. Dampers just to tame peaks.
dhruvmeena96
Headphoneus Supremus
Suggestion 1
Add 20ohms to each 2389 and then series it
Suggestion 2
Add a big ass 47 or 33uF cap
Suggestion 3
Add green to red damper(depends how much you want on 12mm tube
Suggestion 4
38D1XJ007MI/8A
Series it
Lpad it(4.7uF series and parallel)
Increase series resistor to control the amount of bass
Aldo40
Member of the Trade: Alambic-Ears
Played around with this Sonion 20x53 mystery driver, carefully filed a corner to vent it, now that's looking pretty good for about 7 bucks a pair... no changes between tests other than the vent (and the tube might have moved a little in the tack.)
Pic:
For less peak, add one or two yellow damper at the end of the tube. A smaller tube in diameter 0.5_0.2mm can also do the same thing, will be more tense but the efficiency in the sub will be a little less good at listening. With a 1mm tube it is a good compromise. 0.5 or 0.2mm is efficient but I find the efficiency less good. It can be interesting to use it in a low config only for example on a ci22955 to tighten it and with a dynamic driver in SUB configuration. This would energize the sub-bass ensemble because everyone has their flaw and advantages But you have to experiment
For less peak, add one or two yellow damper at the end of the tube. A smaller tube in diameter 0.5_0.2mm can also do the same thing, will be more tense but the efficiency in the sub will be a little less good at listening. With a 1mm tube it is a good compromise. 0.5 or 0.2mm is efficient but I find the efficiency less good. It can be interesting to use it in a low config only for example on a ci22955 to tighten it and with a dynamic driver in SUB configuration. This would energize the sub-bass ensemble because everyone has their flaw and advantages But you have to experiment
That was actually with a 1.5mm id tube (driver has a weird plastic nozzle) but that 1.5k peak leads into the pinna gain for whatever mid driver is used, so I kinda like the way that curve looks. But plenty of room to lownpass with smaller tubing, shouldn't need any dampers there.
I got leads soldered and will play around with a breadboard, probably going with the vented sonion 20x53 and a knowles twfk 33803 for next project. Maybe a dip switch with series resistors on the sonion to cut down on the bass, because that is a strong woofer for sure, but I want to make something with stupid bass so this will work.
Wrapping up my CI-30120 and SR-31843 now. Packaging was... a challenge. Had to abandon the brass tube so mids will probably be flat/elevated, but I can fit a short section of the tube in as an orifice if I need to. Relying on liberal use of B7000 to seal all that, so we'll see how it graphs once it dries.
Verify L pad components? Should be resistors not caps, and parallel resistor is much smaller than series resistor, not same value. Right?
What benefits does L pad have vs series resistor, other than needing lower value resistors and not increasing DCR/impedance too much?
dhruvmeena96
Headphoneus Supremus
lpad is series and parallel resistor mate
driver
parallel resistor
series resistor
2pin or mmcx
it flattens impedance and its more calculative in dB loss
as the above 4.7ohm series parallel Lpad will eat 10dB exact, where ever you put it
All L pad calculators I've seen have a much bigger series resistor, and a small parallel resistor. You said 4.7uf, capacitance, so both those things confused me. You mean L pad with 4.7 ohm series and 4.7 ohm parallel?
And being selfish, for my project I'm thinking of dip switches that I can use to short series resistors to increase bass, so a parallel resistor in an L pad complicates that a bit, unless I use a switch to kill it, then the switching is not intuitive, not ideal but that's ok.
So, same question, benefits of L pad vs series resistors? Edit: you did answer that, thanks, impedance and more exact db reduction.
Last edited:
dhruvmeena96
Headphoneus Supremus
I said 47uF not 4.7uF
And it's for different drivet
Aldo40
Member of the Trade: Alambic-Ears
The purpose of the LPAD is to keep a linear impedance constant.
A single serial resistance will vary the height of the driver’s impedance, but not its non-linearity.
So the resistance in// is there to maintain a linear impedance, regardless of the value used on the serial resistance.
The Lpad linearizes the impedance but does not block its order of magnitude, the higher your serial resistance, the higher the impedance will be.
Depending on the target value of the impedance sought, the two R values may be played.
an example:
With a LPAD 4R7 series and 4R7 //, we will have an impedance around 11 ohms on a 38dx1.
With a LPAD 8R2 series and 8R2//, we will have an impedance around 8 ohms for an identical output(db), so there is a choice to make also depending on the performance of the other drivers used to fine-tune to the best.
The LPAD proposed by Dhruv is correct and is a good starting point
Aldo
Last edited:
Last edited:
Aldo40
Member of the Trade: Alambic-Ears
The 38DX1 can be used without filter if you use the tube and the appropriate damper
Regarding the LPAD on the 38DX1 here is what it gives:
Regarding the LPAD on the 38DX1 here is what it gives:
I can do the math if needed, but could someone post the simplified L pad formula where both resistors are the same value? All the calculators I'm finding have different R1 and R2 values.
Well, here's how the CI-30120 and SR-31843 ended up. I removed the tubes from the CI because the response was a little weird, and turns out the midrange dips by itself so I don't need any physical low pass at all. Both drivers are just playing into open space, all sealed with B7000. Lacking top end for sure, but its another project completed.
This was the CI with the SR inside the zobel (wired reverse phase) and SR also has zobel. Will take me some time to evaluate how it sounds because I'm spoiled by listening to the planars I made, but sounds decent so far.
Could even out the peaks with some foam in the nozzle, but don't think its necessary.
This was the CI with the SR inside the zobel (wired reverse phase) and SR also has zobel. Will take me some time to evaluate how it sounds because I'm spoiled by listening to the planars I made, but sounds decent so far.
Could even out the peaks with some foam in the nozzle, but don't think its necessary.
Last edited:
