If you guys are game for cap rolling, wanna try *no caps"? How?
Rewire your cans for balanced drive using a 3.5mm to 3.5mm stereo cable cut in half. Wire tip to positive and ring to negative. Leave sleeve unconnected. One for each side. Easy to do with cans that have four wires coming out (not 3! cannot work if shared ground). The Status OB-1's are perfect for this as they sound great, have 4 wires, and are cheap so i don't have the fear of ruining $300 cans.
Ok, so now you need to have matched FETs on your amp so that bias is the same on left and right, that is, a DMM will show 6.85v and 6.85v. If off a little, add maybe 220k on top of nominal circa 47R value on R4 on channel that is too high to bring down the bias 0.01v. Do this until matched as close as possible.
Now remove your output caps and jumper with solid wire.
Now get a DAC with balanced output (like a Focusrite 2i4) and use a 1/4in TRS to 3.5mm stereo adapter and connect one amp each to the DAC balanced output. Because the amps have matched bias between channels, the positive and negative phase drive voltages will be identical, hence zero difference between them.
Now you have SE Class A balanced DC drive on the outputs. Your input will still have caps, but that's a much smaller value and we can can use some nice 10uF Silmic plus 1uF Wima there or a big paper-in-oil 10uF cap, etc.
I bet the dynamics and resolution would be fun to hear. Be careful though as an imbalance will subject your cans to about half-rail (survivable for most big over-ear phones, maybe not 116dB IEM's).
Any takers? I know many of you guys have two amps and a spare set of beater cans and can do this
Rewire your cans for balanced drive using a 3.5mm to 3.5mm stereo cable cut in half. Wire tip to positive and ring to negative. Leave sleeve unconnected. One for each side. Easy to do with cans that have four wires coming out (not 3! cannot work if shared ground). The Status OB-1's are perfect for this as they sound great, have 4 wires, and are cheap so i don't have the fear of ruining $300 cans.
Ok, so now you need to have matched FETs on your amp so that bias is the same on left and right, that is, a DMM will show 6.85v and 6.85v. If off a little, add maybe 220k on top of nominal circa 47R value on R4 on channel that is too high to bring down the bias 0.01v. Do this until matched as close as possible.
Now remove your output caps and jumper with solid wire.
Now get a DAC with balanced output (like a Focusrite 2i4) and use a 1/4in TRS to 3.5mm stereo adapter and connect one amp each to the DAC balanced output. Because the amps have matched bias between channels, the positive and negative phase drive voltages will be identical, hence zero difference between them.Now you have SE Class A balanced DC drive on the outputs. Your input will still have caps, but that's a much smaller value and we can can use some nice 10uF Silmic plus 1uF Wima there or a big paper-in-oil 10uF cap, etc.
I bet the dynamics and resolution would be fun to hear. Be careful though as an imbalance will subject your cans to about half-rail (survivable for most big over-ear phones, maybe not 116dB IEM's).
Any takers? I know many of you guys have two amps and a spare set of beater cans and can do this
If you guys are game for cap rolling, wanna try *no caps"? How?
Rewire your cans for balanced drive using a 3.5mm to 3.5mm stereo cable cut in half. Wire tip to positive and ring to negative. Leave sleeve unconnected. One for each side. Easy to do with cans that have four wires coming out (not 3! cannot work if shared ground). The Status OB-1's are perfect for this as they sound great, have 4 wires, and are cheap so i don't have the fear of ruining $300 cans.
Ok, so now you need to have matched FETs on your amp so that bias is the same on left and right, that is, a DMM will show 6.85v and 6.85v. If off a little, add maybe 220k on top of nominal circa 47R value on R4 on channel that is too high to bring down the bias 0.01v. Do this until matched as close as possible.
Now remove your output caps and jumper with solid wire.
Now you have SE Class A balanced DC drive on the outputs. Your input will still have caps, but that's a much smaller value and we can can use some nice 10uF Silmic plus 1uF Wima there or a big paper-in-oil 10uF cap, etc.
I bet the dynamics and resolution would be fun to hear. Be careful though as an imbalance will subject your cans to about half-rail (survivable for most big over-ear phones, maybe not 116dB IEM's).
Any takers? I know many of you guys have two amps and a spare set of beater cans and can do this
I think the challenge is not everyone have balanced drive DAC
I guess I should build my single to balanced phase converter for you guys.
Btw, another nice review of this amp on a pairing with HD-6XX headphones by Gwompki on Head-Fi:
https://www.head-fi.org/f/threads/n...warmer-by-xrk971.833106/page-26#post-13588447
Btw, another nice review of this amp on a pairing with HD-6XX headphones by Gwompki on Head-Fi:
https://www.head-fi.org/f/threads/n...warmer-by-xrk971.833106/page-26#post-13588447
Hi xrk971,
I tried bench mine today into REW and TDA_APL and there is quite some distortion going on compared to direct soundcard loop, had a 50 ohms resistor on output as a virtual head phone load is that alright, oh and at same bench discovered amp is inverting did you know this because that would mean its most correct to let player invert to get cone move in same direction as other systems do.
Few days ago got a pair of OB-1 and so far not so pleased with their sound in for me it sounds like system now have a "Loudnees" button somewhere turned on all the time, that said we all probably have a bit different taste and head phones was not expensive so maybe i try later mod their cable to unpractical solid core because its often my favour and for tonal balance can try some filters into player.
By the way just so you know there is some planned downtime on diyA later today http://www.diyaudio.com/forums/site...iyaudio-will-down-sunday-monday-upgrades.html
I tried bench mine today into REW and TDA_APL and there is quite some distortion going on compared to direct soundcard loop, had a 50 ohms resistor on output as a virtual head phone load is that alright, oh and at same bench discovered amp is inverting did you know this because that would mean its most correct to let player invert to get cone move in same direction as other systems do.
Few days ago got a pair of OB-1 and so far not so pleased with their sound in for me it sounds like system now have a "Loudnees" button somewhere turned on all the time, that said we all probably have a bit different taste and head phones was not expensive so maybe i try later mod their cable to unpractical solid core because its often my favour and for tonal balance can try some filters into player.
By the way just so you know there is some planned downtime on diyA later today http://www.diyaudio.com/forums/site...iyaudio-will-down-sunday-monday-upgrades.html
Hi Byrtt,
The amp was designed more for 250ohm headphones, so 50ohms may see more distortion. Although OB-1's measure close to 70ohms per Innerfidelity, to me, they still sound good. My measurements were done with the self contained 270ohm load resistor on-board and no external resistors. It measured about 0.05%THD driving 700mV RMS into 270ohms using RMAA software. Noise floor was about -126dB (limited by noise floor of Focusrite). Yes, I know that the amp inverts - a common thing with source follower SE topology that we accept for the simplicity of the electronics behind the sound.
Sorry you didn't like sound of OB-1, are you saying it sounds to bass heavy? You probably would prefer sound of DT-880's - more neutral in bass tones.
If you got them through Amazon they are pretty good with returns.
Last edited:
Thanks info then for Christmas i wish me some high impedance head phones if santa can find any
No don't be sorry OB-1 it was my own decision and hardware quality feels very good compared a relative low investment, know they have returns but cost of return shipping compared investment makes me keep them and as said one can play around with them change their cables signature and also its probably possible linearize/compensate their resonance plus inductance and hear how that sounds plus when using a digital player output can be EQed.
Yes for me they are too bass heavy or hot in low region and feel boost starts up around 200Hz and below. Wonder if we listen to music in same way since we have different view listening to same headphones plus amp, my player system is always setup so as DACs see native sample rate so as reconstruction filter is right whether its high resolution or normal CD tracks, do you listen same way or can there be some up/down sample or mismatch in your chain that makes a difference what we base experience on.
The Innerfidelity measurements with a head mannequin shows about +1dB shelf from 200Hz downwards - but not an excessive boost. To me, feels almost like the classic response we like where response is tilted to the right to fall maybe 2-3dB from 40Hz to 20kHz.
https://www.innerfidelity.com/images/StatusSMOB1.pdf
https://www.innerfidelity.com/images/StatusSMOB1.pdf
Great thanks didn't know that source sharing nice data.
Could imagine head phones have another ideal curve so as to be natural singing for human ear which it other curve than for speakers and that i have to study deeper about then, but a comment for curve is it looks to be very hot below 1kHz.
Based on average grey curves one could trace them to use for making DSP filters.
Attachments
Last edited:
I think you are right - relative to the highs, the bass is very hot. Compared to the Sennheiser HD-600 which has been considered a studio quality neutral monitor for 28 years running.
Woa, I did not know that if you link a PDF inside an image it shows it as an image without clicking.
Here is the PDF:
https://www.innerfidelity.com/images/SennheiserHD600.pdf
vs. Beyerdynamic DT880-250's that I have:
https://www.innerfidelity.com/images/BeyerdynamicDT880250ohm.pdf
An externally hosted image should be here but it was not working when we last tested it.
Woa, I did not know that if you link a PDF inside an image it shows it as an image without clicking.
Here is the PDF:
https://www.innerfidelity.com/images/SennheiserHD600.pdf
vs. Beyerdynamic DT880-250's that I have:
An externally hosted image should be here but it was not working when we last tested it.
https://www.innerfidelity.com/images/BeyerdynamicDT880250ohm.pdf
Last edited:
Also got my OB-1's a few days ago and also a bit disappointed in the sound initially. They sound very good on my PCA amp, but crappy on my tube amp. The OB-1 definitely has authority in the mid bass and lower - tight, accurate and goes really low and I love them for that. But they lack detail and clarity in the treble region compared to my AKG's. Guess it is like speakers - they all have different personalities.
But I do not regret buying them. Excellent build quality for $55.
Agree last line and will keep them for later tweaks.
Atmosphere generator at very lows is there and actual very smooth from the data but up here they can't create a real realistic kick drum slam that will need low end to function right with higher harmonics and probably needs both amplitude slope and timing being right as a rhythms thing, with realistic i mean get into recording room and for close mic one get to sense being close to thumbs created from strings kick pedals or sticks, my speaker has this right rhythms thing because nowadays its possible to measure and correct with DSP but after xrk971 shared curves from that site now also have some data to base realistic correction for OB-1
Last edited:
It doesn't work on Chrome either in Win7. Only works on my iPhone.
So Byrtt, you are going to do an FIR convolution to map the OB-1 response to a DT880? You can do that? I never thought about doing it for a headphone. I suppose if you made your own binaural-mic'd head mannequin you could fix any headphone. What I don't understand is why all headphones have the falloff above 10k? It must be the acoustics of the narrow human ear canal? Acts as a TL with resonances and dips and peaks? I don't think you want the response of the in-ear mic to be flat, it needs to follow the natural curve of the ear canal?
Year it should be possible but because there is real resonances taking place inside acoustics we probably can't do all we wan't with pure electric EQ without some other zing penalty here and there but have hope to get better tonality compromise for OB-1, for example the dip at 6kHz is 20dB down relative to 1khz and if that is filled up too heavy think will not get real natural and probably invite other problems. Maybe someone have smarter ways to do it but as a first try i would trace curves from those nice quality pdf files and say as example we want OB-1 to perform as HD 650 i put those two curves at "All SPL" tab in REW and do the math to get the difference, then over in Rephase its possible to EQ that difference flat and if needed also inverted, then it can be saved as wav file to use in convolution engine or one can copy settings for all the filters to ones digital player that could be JRiver, not shure we need any FIR engine here because we know nothing about phase and also most devices direct sound is IIR (minimum phase) so most players will do in they often have IIR correction build in. Have a lot of other stuff to do but will get back within a week about result and share correction filters if anybody interested.
Agree and also don't understand the fall off above 10kHz but in a study into this field probably will take quite some time to be good at and understand right prefer to start get OB-1 response bit closer to the famous ones in steps instead of doing some misunderstand overmuch correction.
EQ filters for Status OB-1 head phones owners that will transfer average of OB-1 grey response curves at innerfidelity website to so called Harman target response curve also found over there.
Prepare setting these filters result in shocking fantastic sound performance and great thanks xrk971 pointing to that website and to the website for sharing those curves making this correction possible.
Black is Harman target response curve, brown is average of OB-1 raw curves measured over there, orange is correction curve when setting below filters, and green is brown times orange showing fair close coherence to black.
HS
-----------------
frq: Q: dB:
-----------------
19900 5 +2,37
17000 2 -3,11
LS
-----------------
frq: Q: dB:
-----------------
10 0,5 +12,00
20 0,5 +7,78
40 0,5 +4,03
80 0,5 +3,18
273 1 -2,07
1450 1 -5,93
PEQ
-----------------
frq: Q: dB:
-----------------
2144 2 +1,48
3266 2 -1,93
4726 4 +4,00
9900 6 -10,37
8300 6 +6,81
5031 7 -0,74
5479 3 +4,74
3743 3 -1,78
4594 4 +1,63
5356 4 -1,63
6104 5 +1,63
8112 5 -1,19
9842 5 -1,33
11507 11 +1,48
14082 6 +2,07
18292 8 +1,19
15330 7 +1,63
Prepare setting these filters result in shocking fantastic sound performance and great thanks xrk971 pointing to that website and to the website for sharing those curves making this correction possible.
Black is Harman target response curve, brown is average of OB-1 raw curves measured over there, orange is correction curve when setting below filters, and green is brown times orange showing fair close coherence to black.
HS
-----------------
frq: Q: dB:
-----------------
19900 5 +2,37
17000 2 -3,11
LS
-----------------
frq: Q: dB:
-----------------
10 0,5 +12,00
20 0,5 +7,78
40 0,5 +4,03
80 0,5 +3,18
273 1 -2,07
1450 1 -5,93
PEQ
-----------------
frq: Q: dB:
-----------------
2144 2 +1,48
3266 2 -1,93
4726 4 +4,00
9900 6 -10,37
8300 6 +6,81
5031 7 -0,74
5479 3 +4,74
3743 3 -1,78
4594 4 +1,63
5356 4 -1,63
6104 5 +1,63
8112 5 -1,19
9842 5 -1,33
11507 11 +1,48
14082 6 +2,07
18292 8 +1,19
15330 7 +1,63
Attachments
Nice!!!
Byrtt, thank you for introducing us to convolution methods as applied to headphones. Are you saying it sounds like listening to something like the Two Towers, almost?
Since I have a soundcard now, albeit only as Focusrite 2i4 USB. I suppose I could do this if shown how.
Or are these simply IIR filters? Simply use something like Jriver?
Very cool!
Byrtt, thank you for introducing us to convolution methods as applied to headphones. Are you saying it sounds like listening to something like the Two Towers, almost?
Since I have a soundcard now, albeit only as Focusrite 2i4 USB. I suppose I could do this if shown how.
Or are these simply IIR filters? Simply use something like Jriver?
Very cool!
Last edited:
Yes it sounds fantastic and pretty shure you will agree, i'm on the second pair of nine volters since that EQ was setup into player in can't stop listening, and yes when batteries are fresh on voltage level its as good as sweet spot in front of Two Towers, try yourself live track "Hotel California" from album "Hell Freezes Over" and dream back sitting in Ronalds couch. Have idea and experience from HD 280 Pro than when changing multi core Litz cable to unpractical solid core it would be even better natural sounding killing the sometimes little hard edges on harmonics making lows more realistic which often is a little syntetic in head phones compared to good tweaked speakers, of course there is no guarantee cable mod will work but will try it later when have a plan how to dismantle and check so its possible to restore with original cable again if needed.
Technically because its pure IIR correction didn't create it as one filter inside wav file to be loaded into players convolution engine because then it gets more complicated to set up in needs of wav file at all possible sample rates and a config file that points to those wav files, for some its a piece of cake to setup but think every one can find out to copy number from those 25 filters above plus probably also good idea when at it also set a filter than reverse polarity output to correct inverting source follower SE topology. In Jriver its pratical have all these filters set in one container then its easy tick on that container and player is armed with correction listening OB-1 via PCA, later untick container and player is ready sending un EQed clean signal to speaker system or other device.
Hope post help to set it up in picture below you see JRiver DSP and over at right is all necessary filters ticked on so they form orange specific correction curve to OB-1 so as Harman target response curve is formed, at left see its one un tick point and all correction inside that PEQ container is disabled, by the way in visual i'm listening 13dB DR Fleetwood Mac track "Never going back" but sadly can hear second set batteries are running cold now so need to go out shopping more elements. It was great you pointed to that site because without those curves didn't know what important to correct for and how a target response for ear canal looks, try it out and if successful request DT880 correction too or a detailed step by step guide.
Technically because its pure IIR correction didn't create it as one filter inside wav file to be loaded into players convolution engine because then it gets more complicated to set up in needs of wav file at all possible sample rates and a config file that points to those wav files, for some its a piece of cake to setup but think every one can find out to copy number from those 25 filters above plus probably also good idea when at it also set a filter than reverse polarity output to correct inverting source follower SE topology. In Jriver its pratical have all these filters set in one container then its easy tick on that container and player is armed with correction listening OB-1 via PCA, later untick container and player is ready sending un EQed clean signal to speaker system or other device.
Hope post help to set it up in picture below you see JRiver DSP and over at right is all necessary filters ticked on so they form orange specific correction curve to OB-1 so as Harman target response curve is formed, at left see its one un tick point and all correction inside that PEQ container is disabled, by the way in visual i'm listening 13dB DR Fleetwood Mac track "Never going back" but sadly can hear second set batteries are running cold now so need to go out shopping more elements. It was great you pointed to that site because without those curves didn't know what important to correct for and how a target response for ear canal looks, try it out and if successful request DT880 correction too or a detailed step by step guide.
Attachments
Byrtt,
Thank you so much for making this Easy.
I am going to buy a copy of Jriver (finally). Can't wait to hear this in balanced drive mode on OB-1's. If you say it's like sitting at sweet spot on couch in Assen listening to the Two Towers - this is indeed special treat awaiting for me.
You need rechargeable Li-ions, here is what I use:
https://www.amazon.com/EBL-Self-Discharge-Lithium-ion-Rechargeable-Batteries/dp/B00EQ3U2AA
You can charge a set while listening to other set. Also, if you plan on using mostly for desktop use, I can show you how to make a desktop PSU so that regular 5.5mm 12v SMPS can power amp. I use DC step up to 18v (1.2MHz switcher) and CRCRC filter.
I thought Litz wire is the best? That's what is in the current OB-1's and my DT880's. Although I see a lot of boutique headphone audio stores selling outrageously priced hand made cables using some sort of hand-twisted pair clear (pink looking copper) 3.5mm to 3.5mm (3inch long) cables for the cost of my amp! What is the world coming to?
$265:
https://www.aloaudio.com/shop/reference-8-mini-to-mini/
Thank you so much for making this Easy.
I am going to buy a copy of Jriver (finally). Can't wait to hear this in balanced drive mode on OB-1's. If you say it's like sitting at sweet spot on couch in Assen listening to the Two Towers - this is indeed special treat awaiting for me.
You need rechargeable Li-ions, here is what I use:
https://www.amazon.com/EBL-Self-Discharge-Lithium-ion-Rechargeable-Batteries/dp/B00EQ3U2AA
You can charge a set while listening to other set. Also, if you plan on using mostly for desktop use, I can show you how to make a desktop PSU so that regular 5.5mm 12v SMPS can power amp. I use DC step up to 18v (1.2MHz switcher) and CRCRC filter.
An externally hosted image should be here but it was not working when we last tested it.
I thought Litz wire is the best? That's what is in the current OB-1's and my DT880's. Although I see a lot of boutique headphone audio stores selling outrageously priced hand made cables using some sort of hand-twisted pair clear (pink looking copper) 3.5mm to 3.5mm (3inch long) cables for the cost of my amp! What is the world coming to?
$265:
https://www.aloaudio.com/shop/reference-8-mini-to-mini/
Last edited:
DSP
Yes, this is very exciting and all new to me. Still not sure how to implement this, but quess I will have to start with installing JRiver. I am currently using MusicBee, which can import WinAmp DSP plugins - which might or might not work according to Wiki. Goodbye MusicBee.
Byrtt,
Thank you so much for making this Easy.
I am going to buy a copy of Jriver (finally). Can't wait to hear this in balanced drive mode on OB-1's. If you say it's like sitting at sweet spot on couch in Assen listening to the Two Towers - this is indeed special treat awaiting for me.
Yes, this is very exciting and all new to me. Still not sure how to implement this, but quess I will have to start with installing JRiver. I am currently using MusicBee, which can import WinAmp DSP plugins - which might or might not work according to Wiki. Goodbye MusicBee.