Have stumbled upon an interesting issue with the headphone impedance vs. the amp used to drive them. A lot of planar headphones have very low impedance (less than 30 ohms). The planar headphones I currently own (Verum 2) sound great with when driven by the Topping D900/A900 stack. They lose some fidelity when using a OTL Tube amp or a Canever ZeroUno DAC/HPA. The headphone amp in the ZeroUno is great, a class A zero feedback amp, BUT, is designed to drive headphones from 30 ohms to 600 ohms. Any headphone that meets this spec this sounds wonderful. The headphones that have lower impedance than 30 ohms seem to lose just a bit of optimum fidelity when using this amp. My OTL headphone amp was designed much the same, in that it was designed to drive headphones from 32 ohms to 600 ohms.
I think the impedance match between a given headphone and headphone amp is a subject that reviews need to pay more attention to.
My very old high-school physics recollection is all but failing me here, but I read this as basically saying that without a transformer (e.g. OTL amp), you want a very large impedance at the āreceiving device,ā i.e. your headphones.
A few years ago, I first tested my freshly built Bottlehead Crack OTL amp with cheaper headphones. I feared that my amateur soldering skills would cause a short or blow up, and started with the 150 ohm HD-58x. Bad idea. Scratchy, thin, and very unpleasant. I switched to the HD-6xx and it was fine. The BHC sounds decent with 300 or 600 ohm headphones.
My OTL headphone amp is based on a design from an engineer who posted an article in Vacuum Tube Valley (VTV) back in the mid 90ās. He called it a Kludge, and provided the schematic.
It uses a 6SL7 input tube, and a 6080 output tube. He specifically designed it to work with Grado cans (32 ohms) as well as the high impedance cans from Beyer, Sennheiser, etc.
When I briefly inquired on SBAF about the suitability of my headphone amp (DSHA-3FN) with the ETA Uli (16 ohm dynamic), the amp designer himself said electrically it would be fine, but Marv mentioned the damping factor might be too low. The amp has a 4 ohm output impedance, so 16/4 not ideal.
In any case OTL tube amps are not generally going to sound good with low impedance headphones, though there are some exceptions. Schiit uses an āimpedance multiplierā switch in Folkvangr and Valhalla 3 to help deal with low impedance headphones.
I donāt understand all the technical jargon, or know enough about OTL amps to comment on that. Generally speaking though, the load impedance should be higher than the source impedance for the best audio connections. (Some audiophiles will argue this point though.)
When you connect wired headphones to an amp, the amp is the source, and the headphone is the load. So you want the input impedance on the headphone to be higher than the output impedance on the amp for the best connection. I believe this maximizes the signal and voltage transfer. And it also improves the electrical damping on the transducerās drivers, reducing unwanted movement (ringing or distortion). Thatās the theory anyway. And it fits pretty well with my own personal experience with higher impedance amps in the 25 to 50 ohm range.
This doesnāt mean that higher impedance amps are bad. Just that they need to be paired with even higher impedance headphones for the best signal and sound quality. Some audiophiles donāt seem to mind the loss in clarity and increase in distortion with a higher impedance amp and lower damp factor though. And may even prefer it.
Another factor to consider is the relationship between the headphoneās sensitivity and the ampās gain.
If the lower impedance headphone is also highly sensitive (read loud), and you have to turn the volume way down on the amp for comfortable listening, then you may get more audible noise coming through from the amp as well. Some amps will have multiple gain settings to accomodate a broader range of headphone sensitivities though.
Louder, higher sensitivity headphones will increase the potential audibility of source-based noise though. And simply turning volume down on the amp usually wonāt fix that. Thatās why you also need an amp with an appropriate amount of gain for your headphones. Or a headphone with appropriate sensitivity (ie loudness or quietness) for your amp.
The OTL headphone amp I have is essentially the same as the one GR Research sells on their website. The only difference is mine has a laminate transformer and a less fancy chassis. All the internals are the same. The builder really beefed up the power supply, used all high end film caps, non inductive resistors, separate windings for the tube filaments that are DC vice AC.
Iām using a RCA Red Base 5691, and a Bendix 6080 tube. The amp sounds great with my Sennheiser 600 series headphones, as well as the Ultrasone 15 Veritas. The amp with the Ultrasone headphones are especially noteworthy, as there is great synergy with this combination.
The issue is probably caused by one or more of the following factors:
ā¢ Tube amps prioritize tonal character over strict accuracy, and their linearity is inherently weaker ā especially when they use little or no negative feedback.
ā¢ Class-A amps ā being low-efficiency designs ā can get heavily stressed by low-impedance planar magnetic headphones, which require a lot of current. Based on the manual, that amp can drive them, but it probably doesnāt have much headroom.
ā¢ Balanced drive requires the + and ā sides of the circuit to be perfectly symmetrical, and any mismatch increases differential errors. With vacuum tubes, itās unrealistic to expect perfectly matched pairs, so some imbalance is inevitable.
My guess is that the second factor is the biggest one.
But realistically, itās rarely just a single cause ā itās usually a combination of multiple factors, possibly including things like hysteresis effects or thermal behavior under load.
