Showing 1 of 449 conversations about:
View Full Discussion It took me awhile to notice that something is wrong with my O2+SDAC. My particular unit came with wrong gain configuration (2.5X and 6.5X* instead of 1X and 3.3X listed on the product page) and other units might be affected by this as well (at least from what I've noticed on a teardown from Reddit).
*O2's input stage will overload when set to high gain of 6.5X with SDAC's output at max, which may cause the sound to become distorted but it won't harm the O2.
*Before disassembly "...wait at least 15 seconds with the power turned ON(in) after the power is removed to let the capacitors discharge."-NwAvGuy
*turn CCW to unscrew any of the screw or nut
-Remove the volume knob by removing the hex screw securing it to the pot's shaft.
-Remove the nut securing the potentiometer to the front panel.
-Remove 4 Phillips screws on the front.
by now the front panel should be able to be come off with the jack attached and you should be able to see if the resistors are of right values (R19, R23) and left out(R17, R21) if they aren't contact MD's support.
*the nut is on the jack assembly 1K resistors removed from R17, R21 on the side
-Remove the 1/4" HP jack by turning the black plastic nut CCW (it may help to grip both sides of the jack while turning and use a wide elastic band or piece of cloth to help grip and prevent an scratches on the nut).
-Unscrew 4 Phillips screws and 1 Torx screw on the back.
The back panel should come right off.
-Slide PCB out the back of the enclosure (it will get stuck if you slide it out through the front because of how the SDAC is mounted)
THE GAIN RESISTORS
"Resistor Values – The O2’s gain (for one channel) is:
Low Gain Ratio = 1 + R16/R17
High Gain Ratio = 1 + R16/R19
Voltage Gain in dB = 20 * Log(Gain Ratio)
…For 1X gain just leave out the gain resistors"-NwAvGuy
Switch out: R17and R21 needs to be left out for 1X gain (gain switch pushed out) but 1K ohms resistors are installed.
So the gain is 1+1500/1000 = 2.5
Switch in: R19 and R23 needs to be 665 ohms or 649 ohms (closest standard resistor values) but 274 ohms resistors are installed.
665 ohms: 1+(1500/665) = 3.26 (~3.3X)
649 ohms: 1+(1500/649) = 3.31 (~3.3X)
274 ohms: 1+(1500/274) = 6.47 (~6.5X)
*this is done at your own risk
Method 1(no soldering required): Clip off or desolder the "high gain resistors" (R19, R23) and leave the "low gain" resistors intact(R17, R21).
Now the gain switch be the following:
SWITCH IN = LOW GAIN (1X)
SWITCH OUT = HIGH GAIN (2.5X)
Method 2 (soldering required): Do the same as the first method so while you get the resistors you still have high and low gain(1X and 2.5X). Clip out R17, R21 and solder in the "high gain" resistors in R19 and R23.
O2's gain resistors:
in "THE GAIN RESISTORS" section
O2's maximum input:
The teardown post on Reddit:
SDAC measurements on ASR:
Yes, so just clip off the 2 gain resistors R19 and R23 (274 ohms, red violet yellow black brown) near the volume pot and now you'll get 1X with gain switch in and 2.5X with gain switch out.
Thanks for sharing this; I'd been aware of clipping when the switch was pushed in and didn't know why. Have desoldered the 274 Ohm resistors and much happier with the 1x and 2.5x gain setup compared with 2.5x and 6.5x.
Hopefully Massdrop fixes this in their next batch. Hopefully it drops again...
They need to fix this for current owners. Will post in head-fi.
As much as I appreciate the time and effort you put into this, I'm not about to go about opening it, much less making modifications and voiding my warranty.
Is there any other way how to find out the gain configuration without opening the unit?
Yeah sure, except I don't live in the US, but the warranty issue was mainly meant for the modification, as in the EU we have a similar bill either in effect or on its way.
I don't exactly trust myself with this kind of handling electronics, and I'd prefer potentially having a product with a wrong configuration over potentially electrocuting myself, especially by accident.
Hence my question about how can I tell differently.
Im interested to see what Massdrop support has to say. Please update us when you get a reply.
Yes, there is another way to check the gain.
You'll need to measure O2's output voltage with O2 and SDAC's* volume at max using a multimeter or a voltmeter.
*SDAC will need to play back a 60Hz 0dBFS sine wave during testing which you can create one in Audacity (Audacity > Generate > Tone > Sine) since most meters are not as accurate at higher frequencies.
O2's output voltage should measure ~2.15Vrms with gain switch out("low gain") or the same as SDAC's output if the gain is 1X, for 2.5X it will be ~5.38Vrms. As for 6.5X gain the O2 will max out at ~7Vrms.
Vout / Vin = Gain
Generate menu in Audacity:
NwAvGuy on gain(there is a section on measuring the gain of an amplifier):
Hey OpasTanti, thanks for your post! We investigated what you reported and found that this was a batch-specific mixup, impacting a number of purchasers. We've contacted all impacted purchasers and thanks again for sharing this with the community.
I noticed this when I first played music on this amp.
At high gain it distorted.
I was able to drive everything with low gain so it was okay but was thinking why.....
I will try to desolder the resistors and see how it goes.
Now I understand why I get distorsion at high gain. The SDAC’s output at 2V should not be joined to the O2 amp at more than 3.3 gain (no more than 2v x 3.3x gain). Now at high gain it’s at 2v x 6.5x gain. The only alternative is to leave the gain switch at low gain (2.5x) all the time. I am not impressed. And now Massdrop is offering at 15$ voucher... or a return/reimbursement. Since there is a fix, a question of bad match of resistors, and since most of consumers are not electricians, I think the ONLY measure that Massdrop should take in such circumstances is to offer to reship the units to Massdrop free of charge so that Massdrop may repair the units up to its specifications and return them to customers. Massdrop should take care of the problem by reparing the units and send them back to customers free of charge. Don’t tell me you don’t provide the after sale service or have the workforce to do this. Then Massdrop find yourself a third party repair shop to do so for you and take your responsibility. Now it’s a flaw with the Massdrop O2/SDAC amps and you won’t repair them. Next it will be the same for your Massdrop/Cavalli amps: you won’t be able to service them and will again issue stupid vouchers. I am pissed. Pissed. Get your act together Massdrop! Get yourself a repair and service strategy Or else you will build yourself a very bad after-sale reputation. I am formally asking you to take care of the repair, without repair costs and without shipping costs. Please govern yourself accordingly.
Thank you so much for this explanation. I thought there was something up with the distortion on high gain but none of my gear needs it so I just ignored the issue.
As for Massdrop saying only 5 people complained about the problem, I'll admit I didn't complain because I was too new to audio gear to know the distortion I was hearing was a problem with the amp.
Just want to make sure my O2 + SDAC is working the same as everyone else's, when the button is pressed in is louder (high gain setting?) than when the button is out is that right?
I'm a complete electronics noob so bare with me, I've opened mine up and it pretty much looks like what Stephen has, so if i want to fix it i just have to cut/de-solder the two resistors nearest the volume control right?
Thanks for this post.
So can we just follow your advice on swapping the resistors to get 1x (switch out) and 3.3x (switch in) gain?
- R17 and R21 (next to the gain switch) left out
- R19 and R23 (next to the volume knob) 2 x 665 ohms (metal film resistor < 3.9mm, 1% tolerance) .
Or would it be more complicated than just soldering?
This issue is very annoying and I somewhat agree with PeteMtl, although I'm able to do the soldering it's a pain to do. I don't think the $15 voucher is enough compensation especially for international customers who already pay an addition $15 for international shipping + customs tax. I will probably be returning the item for a refund and buy it again when the next drop comes around.
Well for international customers a return is not exactly a good option, as for example I had to pay +20% in tax and customs duty. So returning it, even with no shipping costs, would leave me at least - 30$. So I'll be keeping mine as is, I don't use high gain anyway, but I agree that they should have a better QC overall. This is the 2nd time there's a problem with a Massdrop product I purchased.
@bnShadow13 I was lucky and actually didn't get hit with customs on this, so it shouldn't be a problem for me, returing it. Are you from Canada? They shipped with usps which has a low likelihood of getting hit.
It's also funny how my cheap thin hex screw driver just snapped in half while I was trying to unscrew the screw in the dial. So I have no way of getting it open.
@WarningHPB it should be louder with the gain switch in (high gain), for the fix just snip off the two resistors closest to the potentiometer as you’ve mentioned but now to select low gain(1X) leave the gain switch pushed in and switch pushed out for (2.5X).
@Jiv_au that’s all you need to do for 1X gain (in) and 3.3X gain (out) some soldering and clipping leads, and don’t forget 1% tolerance or better for the resistors.
I'm from Europe, Czech Republic... So... Yeah
From this incident... I am learning that I should not put all my faith in MD.
Yes, 15 dollars is something but this only benefits MD, not the MD customers.
If MD did not offer any compensation then that would have caused an Exodus.
Same thing happened recently with VE Clan Cables.
A complete mess up.
Even that they offered a full refund....
And this O2SDAC a 15 dollars.... hmm
And that is only IF I buy something from MD again...
I joined R2R Dac but now I am thinking should I pull out or not.
Not so much of trust raising events keep repeating...
It would have looked better if MD QA team found this themselves and took proactive measure instead of giving a 15 dollars voucher.
There should have been an effort to fix the units instead of giving out vouchers.
What if I received R2R Dac at the end of this year and same thing happens....
I know I would be given a voucher again.....
Maybe another 15 dollars...
For those who would like to try soldering...
Can you send us the correct resistors to replace?
I would like to make mine same as how the good batch was made like.
So if you can ship the parts free of charge I wish to solder it myself.
Please provide the instructional steps as well on this discussion thread.
And I insist that the product warranty would not be void by having us performing the fix.
With this and the 15 dollars coupon, at least for myself, I would consider this issue resolved.
Edit: Yep. MD must fix it. They must provide free shipping back, fix the flaw and send it back to me free of charge. Until then, this is not resolved and I will not buy anything from MD. Still thinking whether I should cancel R2R DAC.
Thanks for that - 1% toletance or better resistor.
On another note, I've now started asking myself whether I really want 1x gain?
I've been using 2.5x gain (unknowingly) with my 35, 40, and 300ohm HP without much issues, as best as my ears can tell. I would assume 1x gain isn't enough for 300ohm (HD6xx) since I often have it at 12 o'clock volume. 9 o'clock for 35ohm (Fidelio X2).
3.3x gain is something I would like to have, as that is the max power possible without distortion through the SDAC + O2 combination. But do I need it if I don't own, or intend to own, a much higher impedance HP (e.g. 600ohm)? Some people think 7volts (3.3x gain) into 300ohm to produce 120dBSPL is needed, but that isn't for me.
6.5x gain combined with SDAC is the problem (distortion). But I'm wondering if I might need it when using the amp section only (O2 amp-only version comes with 6.5x gain option).
So for me the dillema now is choosing which gain combination I want.
In fact, leaving it AS IS is also an option.
2.5x/3.3x or 3.3x/6.5x or 2.5x/6.5x or 1x/3.3x or 1x/2.5x.
You could solder in SIP sockets for solderless gain change.
NwAvGuy did list one on O2's BOM. (Aries 10-0518-10 )
With products such as the Massdrop O2, SDAC, Cavallis, R2R, Massdrop is not a distributor/reseller/retailer anymore. They are a manufacturer: they offer exclusive products which carry their brand name, even if they use third parties to assemble these products in China. Massdrop is now a manufacturer, just like Apple, Sennheiser, even though they subcontract the actual production steps. As a responsible manufacturer, Massdrop should be responsible for after sale service under warranty, they should take care of repairs and servicing. If their products have manufacturing flaws like in this case, they should govern themselves accordingly and act like a responsible manufacturer. If they do not change their culture of “reseller only” company, which they are not anymore, then they will built themselves a very bad reputation. I have been a good ambassador to their business and I referr to many friends to their business. If they don’t act quickly, I will become a strong detractor instead of a an ambassador. Massdrop has to recall all affected units, repair them and resend them to customers under warranty, without costs. That is the responsible thing to do. Vouchers are a way to say to customers: we don’t care, take that money and shut up. My pending order for the HD58x will be the last item I purchase from them until they put their act together and become a responsible manufacturer.
How can you bear to listen to your HD6XX at 12 o'clock. With my windows volume at 100%, it's already getting fairly loud at 9 o'clock. (12 o'clock would be unbearable) 9 o'clock is good when windows volume is at around 50%.
Hrmm, did I sell of my remaining SIP sockets?
Yes, last year. Fiddlesticks.
That would certainly make me a proud massdropper!!!
It would boost my faith in MD too.
That would surely open up my wallet more and buy with confidence.
I think it has to do with my music collection (classical) - it seems most are way below the maximum so that they don't clip even on 6.5x gain. I had to create a sine wave with amplitude of 1 in Audacity to hear the 6.5x clipping first-hand.
I do turn it down to 10 o'clock volume when on Skype.
@OpasTanti Thanks, I'll look into the SIP sockets.
At 2.1v output, anything over 3.3 miliamperes will overload if the gain is 6.5x. It’s a flaw and the maximum voltage allowed is 7v on the O2. With the SDAC 2.1v max output, the gain is optimal at 3.3x and should not exceed this level. The O2 will overload over 7v (3.3x gain x 2.1v is the maximum voltage permitted by the O2 without distorting = 7v. Massdrop should offer to repair and correct the gain modes to 1x and 3.3x instead of 2.4x and 6.5x. With the SDAC‘s output at 2.1v, anything over 3.3v is a design flaw.
Just $15 voucher lol. They should fix them all. They have to provide what we bought. I cannot trust Massdrop anymore. At this circumstance, it is impossible for me to buy at Massdrop again.
So, for everyone who has gotten back their 1x gain, how good is it with your headphone? Anyone with HD6xx?
Is 1x gain same as listening directly to the SDAC via RCA (e.g. using RCA to 3.5mm stereo adapter).
I suppose there is no warranty since they're just offering to refund.
just want to clarify here, in your original picture it looks like the other pair is clipped?
That is the problem with Massdrop: no warranty, no after sale service, no repair. It’s time for them to change that. they have exclusive products made to their specifications. They are now partly a manufacfacturer (for products like the O2, SDAC, Cavalli amps, R2R DACs), and as such they should provide service, repair and honour the warranties. If they don’t change, they will get a bad reputation and will loose ambassadors and clients. I include myself here.
I won’t buy else from Massdrop until they get their act together. For example, what would happen if I buy a Cavalli Amp And it fails? Cavalli doesn’t exist anymore as a manufacturer. Massdrop is the manufacturer: they bought the licences and produce the amps with subcontractors. What happens if a Massdrop/Cavalli amp doesn’t perform well and needs servicing? Vouchers offered again? That doesn’t make sense. Same for the O2 and O2/SDAC. They must provide a correction to the resistor flaw in manufacturing. A responsible manufacturer would offer to repair and send back free of charge. They may subcontract the service and repairs, but they must offer such after sale service. It doesn’t make sense without it.
I do agree to some extent, though a valid outcome of the resolution of warranties is a full refund which they offered. I think as a company, it doesn't look good not standing behind your products.
They should also be sending warranty cards with what they cover for each product, for example in this case they didn't state they would refund if there are issues. I'm not sure if this is against FTC regulations, but I snipped my resistors at this point
Yes, that was a mistake I made initially so the original picture on that post is a "how to not" and not a "how to do".
Regarding Massdrop's warranty, I'm unable to find any info of it on their terms and conditions page (except for the RETURN POLICY).
Now, are there enough of us DIY folk to make a 665/649 ohm resister GB worth it? I'm a bit short on hobby shops to swing by for local components in my area (that can't possibly be true, can it? I mean, Radio Row died in the 60's and all...)
@Jiv_au with the O2 on 1X gain with my sensitive IEMs and the volume maxed the noise is almost non-existent, I could just barely hear it. Also lower gain = lower noise and distortion.
"Is 1x gain same as listening directly to the SDAC via RCA..."
Even though output voltage will be the same, DAC's output have high output impedance and it might not have enough power to drive headphones from my understanding.
Power, Gain, Noise and Dynamic range :
Not to mention the 3-4 month wait we all went thru, if I recall correctly.
Thanks again. I didn't realise the noise until you mentioned it. But I could only hear it when the volume is at max at 6.5x gain.
To answer my own question: Is 1x gain sufficient for HD6xx?
YMMV if you like your music loud, but I think prolong exposure at 1x gain with max volume would cause hearing damage even with the HD6xx.
Here's my mod. Apologies for the poor picture quality.
(Note: I'm using 680ohm instead of 665 or 649 as I couldn't source them locally)
Finally, 44 weeks after the initial report, I went ahead and did the fix using 680 resistors as per the previous poster and completely removing the low gain resistors, solder and all.
Bit of troubleshooting and a maths question:
My high gain sounds lovely and clean all the way through the volume knob, but my low gain is now giving me some hum at certain points along the knob, with a solid buzz at full volume. I'm wondering if leaving the through holes empty is allowing for some interference (which seems unlikely). Do you think a bit of solder would help? As an alternate solution, I have some 1k5 resistors on hand, if I understand your equations correctly, that would balance out to 1x gain. Is there any reason not to do that?
There shouldn't be any sort of buzz or hum with or without resistors installed, something is wrong. I suppose you could check for soldering defects around the gain resistors (a solder bridge on the gain switch pins?).
If there aren't any then I recommend you to ask over in the O2 thread on diyaudio.com
...yes, there is a reason not to...using a 1500 ohms resistor will give you a gain of 2X not 1X
( 1+(1500/1500) = 1+(1) = 2 )
Solder bridge seems likely, I'll pull out the solder wick and give it a clean up