Home-Made IEMs
Jan 5, 2021 at 10:52 PM Post #12,076 of 16,074
to the measure o of a driver alone, you will not see anything different whether they are reversed or not :wink:
For the meet, yes it would be great but we are so far away, then I already have my IEM company but I am all alone :D
Congrats on your venture. Would you happen to let us know your company? Would love to follow you on social media. :D
 
Jan 5, 2021 at 11:35 PM Post #12,077 of 16,074
Made an impedance jig! After figuring out that the little usb sound card I was using for my imm6 wouldn't work because it only had a mono mic input, I got these results with the computer main soundcard input. Probably messed up the impedance calibration in room eq wizard and can't figure out how to redo it, but oh well, these results seem reasonable.

Knowles SR-31843 with 39 ohm and 4.7uf zobel looks pretty flat to me. 10mm dd from KZ EDX is real flat, as expected. Knowles CI-30120 has a huge spike and exponential increase at 20k, again as expected.

Ignore the left Y scale (no idea?) but impedance is on the right side of the graph. Also ignore my 31020/30120 typo :frowning2: troubleshooting took me a few beers to get through.

Will be playing with it more this weekend. Fun stuff.

impedance.jpg
 
Jan 5, 2021 at 11:48 PM Post #12,078 of 16,074
Apparently the zobel circuit causes impedance to approach the zobel resistor value? So I probably overshot with 39 ohm, and could get the 31843 impedance a little flatter if I used a smaller resistor, 1.25x DC resistance (that exact driver measured 25.8 ohms DCR and 6.0mH, for reference.) So a 33ohm resistor I assume would have been a better choice, and end up with even flatter impedance.

I couldn't find the 31843 impedance curve, but it is a BA so it would have a similar spike/rise to the 30120 shown.
 
Jan 6, 2021 at 12:04 AM Post #12,079 of 16,074
Apparently the zobel circuit causes impedance to approach the zobel resistor value? So I probably overshot with 39 ohm, and could get the 31843 impedance a little flatter if I used a smaller resistor, 1.25x DC resistance (that exact driver measured 25.8 ohms DCR and 6.0mH, for reference.) So a 33ohm resistor I assume would have been a better choice, and end up with even flatter impedance.

I couldn't find the 31843 impedance curve, but it is a BA so it would have a similar spike/rise to the 30120 shown.
not kinda and sort of yes
i mean you wont approach resistor value exactly and thats not the aim as its 1.25x more
and also you can even go lower and higher with resistor changes(warning, extreme value can shift frequency response
but also 39ohm is high for SR

also the inductance on that SR is ~6.6mH
so zobel comes out to be approx 6.8uF and 31ohms
 
Jan 6, 2021 at 12:15 AM Post #12,080 of 16,074
not kinda and sort of yes
i mean you wont approach resistor value exactly and thats not the aim as its 1.25x more
and also you can even go lower and higher with resistor changes(warning, extreme value can shift frequency response
but also 39ohm is high for SR

also the inductance on that SR is ~6.6mH
so zobel comes out to be approx 6.8uF and 31ohms

I measured it at 6.0 mH but have a new LCR meter on the way because I don't trust this one, it was cheap and is probably 10 years old. But your numbers sound closer than what I used.

Also interesting to see on the graph how the impedance flattens out at what would be the high pass frequency of a 4.7uf cap and 39ohm load, ~870hz. Makes sense.
 
Jan 6, 2021 at 1:15 AM Post #12,082 of 16,074
Yes 868 hz in theory.
If you ad resistor the impedance the amplifier sees will be not the same like the resistor you ad .

As i know for a low pass filter for example if you add a 20 ohm resistor in series with a cap in parallel to a 20 ohm speaker then the impedance that amplifier sees will be 40 ohms .

And the opposite is with high pass filter
If you are adding a 20 ohm resistor in parallel with a cap in series to a 20 speaker then the impedance that amplifier seewill be 10 ohms .

Am i right or maybe not ?

I will have all my breadboard components by this weekend, so I can check and let you know :)

I do still need to research RC low pass filters and L pads to understand them better, but I have a few days for that.
 
Jan 6, 2021 at 5:30 AM Post #12,086 of 16,074
I suggest you first read TOS of head-fi
Hello potrius-g,

I am asked a question, I answer it.
TOS=forum rules?
if that’s what this is about
It will be done correctly in the dedicated thread, don’t worry about me it will be done correctly, as I do on TN or elsewhere.
Aldo
 
Last edited:
Jan 6, 2021 at 5:45 AM Post #12,088 of 16,074
Well, I am probably the one that is going to start the thread actually :)

Since I am also in France I have had the privilege of auditioning two models of @Aldo40 the Bali (really shortly) and then spend some significant time with the prototype of the upcoming Mavericks which I have previewed. I'll share more soon on that.
 
Jan 6, 2021 at 6:13 AM Post #12,090 of 16,074
OK, good to hear. Some of the guys from this threat got banned, for posting pro work, so I just wanted to warn you
i mean, thats what i hate
at least let people start a bit, and once they are successful enough
then apply the rule
way more beneficial

but then, i am not starting anything anytime soon
so i am safe
 

Users who are viewing this thread

Back
Top