Today we’re excited to announce our new C5 Headphone Amplifier:
C5 Production Status
- Design: 100% complete
- Engineering Samples: Approved
- Benchmarks: 100% Complete
- PCB Assembly: 100% Complete, on a UPS truck
- Enclosures: In stock, awaiting anodizing February 19-22
Barring unforeseen catastrophic failures, C5 preorders will ship no later than February 28. Review samples and distributor orders will ship on February 22.
Wed., March 13 Update: The C5 preorder sold out this morning. Mass production began two weeks ago, so C5 will be in stock in just 2-4 weeks.
Tues., March 12 Update: 99% of preorders have shipped! New Slate C5’s ship immediately. New Red C5’s will ship next Tuesday, March 19.
Mon., March 11 Update: Slate anodizing remains on schedule for completion tomorrow morning. Our anodizer says this batch is turning out as desired.
Fri., March 8 Update: Over 75% of preorders have shipped (all Red and Silver). All Slate preorders are scheduled to ship by Tuesday evening.
Thurs., March 7 Update: We inspected and approved a new batch of enclosures today, and now expect to complete preorder shipments within 3 business days.
Tues., March 5 Update: We have encountered an “unforeseen catastrophic failure” as I wrote on February 16th. The enclosure delay mentioned last week was caused by our local metal finishing shop. Their job is to apply the smooth, blasted finish to C5 (bead blasting). As of last Tuesday, they realized they’d blasted our C5 cases with the wrong parameters. Already one week late, they called on 2/26 and promised to start over and deliver perfect pieces by Thursday, 2/28. All seemed fine, until we anodized those pieces. Our anodizer called this morning apologetically. His shop follows this blog and is well aware of the significance and beauty of C5. Every single piece from the “fixed” batch was not fixed. Anodizing had revealed severe cosmetic defects–all pieces from 2/28 were junk.
I drove to the blasting facility this afternoon and handed them two C5 enclosures: one from the first batch, and one from the 2/28 batch. The production manager was as shocked as we are, and will begin another batch in the morning with strict supervision.
It’s an absolute shame that one shop foiled our ambitions to ship early. As soon as we realized the possibility of failure, we began identical blasting production at a second metal finishing shop, for redundancy.
The Good News:
- We have a strong supply of raw enclosures; without delays, metal finishing normally takes just 2-5 business days.
- Over 61.5% of C5 preorders have already shipped
For those affected by the delay (mainly Silver and Slate preorders), we promise to upgrade all USPS First Class shipments to USPS Priority. Please do not hesitate to contact us if we can be more helpful!
Mon., March 4 Update: All Red C5’s have shipped. All outstanding preorders will ship
Thurs., Feb 28 Update: Enclosure anodizing is in progress, and we still expect to ship on Monday afternoon.
Wed., Feb 27 Update: A large batch of C5 enclosures will enter anodizing by 1PM Thursday (a 1-2 day process). Preorders will resume shipping on Monday afternoon.
Tues., Feb 26 Update: Over 25% of C5 preorders have shipped. All PCBs have been tested and are awaiting final enclosure assembly. Our enclosure finishing contractors called this morning to report a 2-3 day delay, so remaining C5 orders will ship Feb 28 thru March 6.[/expand]
C5 is not a response to the Objective2, nor to competitors’ products. C5 began as a fun project in 2012 to build a smarter headphone amplifier to solve the single greatest hindrance we see in DIY audio: the analog volume potentiometer.
C5 is built for portable users who need a small amp with USB recharging, exceptionally low noise, sufficient output power, long battery life, and most importantly, a super fine volume control to handle sensitive headphones and IEM’s.
Analog Potentiometers’ True Performance
I interviewed an electrical engineer in January. He’d applied here after discovering the Objective2, and during our discussion, he asked, “What’s there left to design? The O2 is audibly perfect. How can you build anything better than that?”
Easy, there’s no such thing as a perfect product. Every design has its unique goals and constraints.
Our Chinese competitor says the Alps RK097 analog pots are “still the best” solution, presumably based on low cost and decent benchmark performance. NwAvGuy dismissed digital potentiometers as too expensive to implement with good performance, while acknowledging the major shortfall of the Alps pot he settled on for the Objective2:
THE CHANNEL BALANCE PROBLEM: Devices with conventional volume controls may have audible channel imbalance at very low volumes [i.e., one side is much louder than the other –JDS]. It’s extremely difficult to manufacture volume control potentiometers that maintain tight channel balance below about -40 dB (referenced to full volume). — NwAvGuy
In this article he briefly describes how imbalance can be resolved with proper excess gain. The excess gain problem is primarily why we’ve custom built cMoyBBs for over five years. That is, setting a suitable gain usually avoids channel imbalance. Usually–until you’re sitting in a quiet room and want to listen at low volumes, or until you try a set of high efficiency headphones. Even at 1.0x gain, you’ll encounter major channel imbalance from your so-called high performance analog headphone amplifier in such situations.
Let’s take a look at channel balance of c421, measured by the dScope as I slowly turn the knob from maximum to minimum position:
The yellow line represents the left channel; pink represents the right channel. Lines resting exactly on top of each other indicate audibly perfect balance (no deviation in L and R volume).
c421’s Alps RK10J imbalance grows after just -12dB. The problem becomes severe by -22dB, and it’s quite useless by -30dB. If you called or emailed in the past year, you know we didn’t recommended c421 for IEM’s. If you tried c421 without seeking our advice, you either listen above the imbalanced region, or you returned the product.
Here’s an Alps RK097 implemented in the cMoyBB (Objective2 uses the same series potentiometer):
Channel balance of the RK097 fairs much better, due to its larger mechanical size. You can see 1-2dB deviations at -25dB, with otherwise decent balance down to -40 to -50dB. And that’s where things get nasty. I only managed to turn the knob at a single point when the dScope cycled to measure the imbalanced region, which is exactly the problem IEM users face. You can either turn the knob to mute, or to a point of imbalance, or to a level louder than you’d prefer.
Even with the RK097, we still receive a few emails and phone calls each month about channel imbalance. So what good is a HiFi amplifier when it actually hurts your listening experience?
For years, audiophiles feared digital volume controls. Software based digital volume control is the worst, causing you to “lose bits”–the audio signal itself is digitally divided and becomes less precise. Bad!
Early digital potentiometers solved the basic mechanical problems of analog potentiometers. A digital potentiometer is electrically equivalent to an analog potentiometer. Both feature High, Low, and Wiper terminals. The potentiometer is mechanically or digitally set to determine the ratio of High:Low resistance. However, bad digital pots added a large amount of capacitance to the Wiper, and thus, audiophiles frowned at the resulting THD+N (often 0.1% or worse).
It’s 2013, and it’s finally time to say goodbye to the analog potentiometer. C5 features 64 steps of audibly perfect digital attenuation:
C5 presents only +/-0.1dB of deviation all the way down to -50dB, and only +/-0.55dB at -60dB! [Yes, you can only see 28 steps here, as I’m manually racing the dScope test duration by making larger volume transitions.]
In other words, C5’s digital attenuation achieves perfect audible balance at volumes -20dB lower than the analog Alps RK097. Remember that audio sounds twice as loud every 10dB, so this is no small improvement.
Reference Level Performance
C5 began as an experiment, and even I was skeptical that we could outmatch c421’s THD+N with a digital potentiometer. Thus, we bought a dScope III and set no project deadline and no budget. We’d either continue shipping c421’s, or continue experimenting.
By January 2013, it was clear that C5 was electrically complete. We quietly put C5 PCB’s into production instead of another batch of c421’s. C5 had not only hit our THD+N goal, it had matched the O2!
|Frequency Response||+/- 0.02 dB|
|THD+N (20-20kHz, 150 Ω)||0.0009%|
|THD+N (20-20kHz, 32 Ω)||0.0045%|
|Crosstalk @ 150 Ω||-67 dB|
|Inter-channel Phase @ 1kHz||+/- 0.01°|
|Channel Balance||+/- 0.55 dB, all volume positions|
|Max Output @ 600Ω||4.146 VRMS|
|Max Output @ 150Ω||3.337 VRMS|
|Max Output @ 32Ω||1.010 VRMS|
|Power Supply||14.0 Vpp|
|Output Impedance||2.2 Ω|
|Battery Run Time||11-14 Hours|
|Charge Time||2 Hrs to 80%, < 4 Hrs to 100%|
|Operating Temp||–40°C to 85°C|
|Operating Humidity||0 to 85% Rel. Humidity|
|Dimensions (excluding switches)||99.5 x 61.5 x 14.0 mm (LxWxH)|
Achieving high output power was not a primary goal of C5. Referring to its design objectives, portable users do not need inordinate amounts of power (note: P = V^2/R); you need run-time and just enough power. C5 has equal output power to that of c421, and we know from measurements and subjective results that c421 and C5 are well suited to driving 90+ dB/mW headphones. IEMs and common 32-250 ohm dynamic sets are okay for C5. It’s not meant for your planar orthodynamics. Therefore, instead of setting unnecessary supply voltage and output power, we set adequate supply voltage for moderately demanding headphones and achieved run-time of 14 hours.
Dual LDO Supply Regulation
We wanted to place LDO’s in c421, but there was no way to control them, and without control, unsynchronized LDO’s produce unsafe turn-on transients. C5 is smarter than c421, so its firmware simultaneously enables the positive and negative LDO’s after the rest of the amp has initialized. You hear only a safe turn-on transient with C5, and reap the benefits of ultra-low supply noise.
- Digital Stepped Attenuation (63 steps + mute)
- Dual Gain: 2.3x or 6.5x, MOSFET controlled
- Bass Boost: +6.5dB @ 80Hz
- 20-Minute Low Battery Indication
- 0.1% Thin Film Resistors
- 1200mAh, 3.7V Li-Ion Battery
- Smart USB Charging
- Ultra Low Noise +/-7V rails
- 3.5mm Input and Output Jacks
- Gold Immersion, 4-Layer Printed Circuit Board
- Atmega168A MCU w/Opensource Arduino Firmware
|Feature||How to Use|
|Volume||Hold volume lever left to decrease volume, or right to increase volume.|
|Gain||Push volume control to toggle high/low|
|Bass Boost||Toggle up for normal audio, toggle down for bass boost|
A single LED conveys all of C5’s behaviors:
|Amplifier State||LED Status|
|Low Battery||Flashing Green|
The C5 proof of concept began as an Arduino Pro with messy wires coupled to a c421: DIY style.
Writing firmware for a digital potentiometer is surely a hurdle for many in the DIY audio community. In pursuit of retiring analog potentiometers in DIY HiFi, we’re releasing C5’s firmware under the CC BY-SA 3.0 license. Note that a 6-pin header and ISP programmer are required to write to C5. Enjoy!