Project

# Title Team Members TA Documents Sponsor
1 Vacuum Tube Amplifier
Bingqian Ye
Qichen Jin
Zhen Qin design_review
final_paper
presentation
presentation
proposal
Team members:
Qichen Jin, NetID: qjin4 (qjin4@illinois.edu)
Bingqian Ye, NetID: bye3 (bye3@illinois.edu)

Introduction and motive
Our goal is to mainly tackle on the amplifier portion of audio system. There are basically two types of amplifiers, solid state amplifier and vacuum amplifier. Since this project will only be three or four mouths, the design of solid state amplifier would involve with too much details on the Operational Amplifiers and dozens of feedback loops in high order, we decide to build vacuum tube amplifier, it is less complicated and more achievable in the short deadline. From the market perspective, there is no budget vacuum amplifier available in the US market. Besides, there is some low cost solid state amplifier available and some of them do their job really well (I.E. Onkyo M-5010). In order to test the performance of our amplifier, we would use a pair of cheap speakers for testing. We bought a pair of full range bookshelves drivers for testing. Since this is our first time dealing with vacuum tube circuit, we will mainly choose classical style circuit design. If we have more time, we would study more complicated vacuum tube circuits such as push-pull, and/or class-AB amplifiers. The details of plans for the vacuum tube amplifier is listed below:

Specification:
Class: Class-A amplifier. Reason: This kind of amplifier is the most simplest form. Though it consumes more power and output less power compare to class-B and class-AB, it might have a lower THD, as well as more friendly for us to design and build. Our designated speakers are bookshelves speakers, so high power is not needed here. We can also avoid any complication for using the cut-off region.
Implementation: Single-end. Reason: Technically a class-A amplifier can be implemented by either single-end or push-pull, but for the usual case and design simplicity, as well as limited bugged, we will use single-end as our system.
Choice of tubes: 6J1*2 as preamp, 6P1*2 as power amp. Reason: 6J1 and 6P1 are pretty cheap tubes that are widely available in Russia and China, and we found this kind of tubes have a typical I-V curve, and manageable voltage requirements (~250V anode, ~6.3V filament). In addition, there are also many successful commercial designs that use 6J1 and 6P1 as part of the amplification circuit.
Wattage: 5 - 10 W per channel for the 4 Ohms speakers, 100 - 300 mW per channel for ~300 Ohms headphones. Depending on the actual quality of the tubes, feedback loops factor, and avoid self-excitement (unstable system), we might adjust the power.

Primary Goals:
Frequency response: 60-18 kHz minimum.
Signal to noise ratio (SNR): ≥ 70 dB, the background noise should be neglectable compared to the sound.
Total harmonic distortion (THD): ≤ 3% @ 1 kHz.
Stereo isolation: ≥ 50 dB @ 1 kHz. The left and right channels need to visualize people about the sound coming from different distance as well as angles rather than just two speakers.
Good clarity and separation of frequencies of sound, for symphony, as least three different instruments can be heard at the exact time.

Other parameters:
Speaker sensitivity: 87 ± 3 dB.
Speaker impedance: 4 Ohms.
Speaker wattage: 15 W maximum.
Speaker frequency response: 50 - 20 kHz.
Headphone sensitivity: 90 dB.
Headphone impedance: 300 ohms.
Headphone wattage: 100 mW.
Headphone frequency response: 50 - 20 kHz.

LED Cube

Michael Lin, Raymond Yeh

LED Cube

Featured Project

LED technology is more advanced and much more efficient than traditional incandescent light bulbs and as such our team decided we wanted to build a device related to LEDs. An LED cube is inherently aesthetically pleasing and ours will be capable of displaying 3D animations and lighting patterns with much increased complexity compared to any 2D display of comparable resolution. Environmental interaction will also be able to control the various lighting effects on the cube. Although our plan is for a visually pleasing cube, our implementation can easily be adapted for more practical applications such as displaying 3D models.