Post by Owen Y on Jun 26, 2017 22:29:17 GMT 12
|
Post by colinf on Jun 27, 2017 8:15:07 GMT 12
Yes, very good frequency response so nfb works better in the treble. Toroidals can't take much DC imbalance though.
AMR-iFi R&D
|
Post by michaelw on Jun 27, 2017 10:59:45 GMT 12
what's "very good frequency response" ?
|
Post by guitardude on Jun 27, 2017 13:42:36 GMT 12
Whats the rest of the tube compliment colin ?
|
Post by colinf on Jun 27, 2017 14:46:40 GMT 12
1w frequency response -1db 1hz to 113khz. I haven't received the transformers yet so I'm yet to measure them for myself. Tube complement 6au6, 6072, 5687, 4 x KT150 per channel.
AMR-iFi R&D
|
Post by guitardude on Jun 27, 2017 16:51:08 GMT 12
SS rectified ?
|
Post by colinf on Jun 27, 2017 16:54:39 GMT 12
Tube rectified.
AMR-iFi R&D
|
Post by guitardude on Jun 30, 2017 8:56:33 GMT 12
Sorry late getting back to this one....
So which one is the rectifier tube ?
Power output is conservative for an octet of KT150 ? Similar to the Jadis JA80 in output I think, whereas the new ARC gets about twice that ?
|
Post by colinf on Jun 30, 2017 10:43:00 GMT 12
I'm using 3 5ar4s in parallel to take the current of 4 KT150s. I'm running them in class A. If they were to run in class AB you could expect 160+ watts out of 4 per channel. But distortion in class A is naturally lower and damping factor higher than class AB.
AMR-iFi R&D
|
Post by guitardude on Jun 30, 2017 10:58:44 GMT 12
Sounds good colin, when did you want to hear what they can really do ?
|
Post by colinf on Jun 30, 2017 11:15:35 GMT 12
You mean when did you want to hear what they can really do on your system?? Alas they aren't ready yet, a few weeks away. And they are for someone... I bought 4 Plitrons though, 2 ready for another one to build.
AMR-iFi R&D
|
Post by guitardude on Jun 30, 2017 13:32:02 GMT 12
No, was only pulling your leg. Have no need of anything more for my system.
|
Post by colinf on Jul 29, 2017 16:34:32 GMT 12
They're alive!! I fired them up for the first time yesterday.
AMR-iFi R&D
|
Post by guitardude on Jul 31, 2017 14:46:24 GMT 12
Mine still work too. Now stop mucking about and get back to finishing off the Ongaku !
|
Post by colinf on Jul 31, 2017 15:25:52 GMT 12
I would if my case maker ever gets around to it!
AMR-iFi R&D
|
Post by guitardude on Aug 1, 2017 8:38:06 GMT 12
at the painters I believe....
|
Post by colinf on Aug 7, 2017 12:06:27 GMT 12
I've been measuring my KT150 monoblocks. They run in triode-connected class A with part self- and part fixed-bias. The output is 55w into 8 ohms and 63w into 4ohms, THD just below clipping 0.5% and the best bit, the frequency response: 6Hz to 125kHz at 50w! The Amplimo toroidal output transformers are excellent. They have such good frequency response that the negative feedback in the circuit operates with next to no phase shift all the way up to about 70kHz! So the 20kHz THD looks much the same as the 1kHz THD does. Try that in a normal valve amp! You can hear it too, the top end is very clear and it shimmers, and I've been hearing resolution on my system that I haven't heard in a while. Four KT150s should ordinarily put out over 150w configured in pentode with fixed bias. I decided to connect the KT150s in this amp in triode for better damping factor (30 for 8ohm speaker in this amp) and part self bias for reliability, as KT150s have a reputation for being prone to bias drift in a fully fixed-bias design. Triode operation means the power output is a lot less than with the pentode or ultralinear connection, but with more sonically benign distortion products. I'm pretty happy with these amps!
AMR-iFi R&D
|
Post by colinf on Aug 7, 2017 15:07:20 GMT 12
Here's a pic of the 50Hz digital generator for the Garrard 301 (and other rim drive TTs like the Lenco) I made 2 years ago. You can vary the frequency in 0.1Hz steps from 43 to 72 Hz, which allows operation for 50Hz and 60Hz turntables. The amp inside is a class D amp connected to a step-up transformer for the TT motor voltage. I run the motor at 190Vac to reduce motor vibration. The generator itself is a microprocessor programmed to generate the sine wave with digital synthesis. Hence I called the unit DDS-1 for direct digital synthesis. I should really make another few of these.... multiple image hosting
AMR-iFi R&D
|
Post by michaelw on Aug 7, 2017 18:50:38 GMT 12
nice ! do you use this box for electronic speed change too ?
|
Post by colinf on Aug 7, 2017 20:05:31 GMT 12
Yes, I set the speed by varying the frequency, and on the 301 with speed control on maximum I need about 49.5Hz. The 301 warms up over about 15mins so the speed is right after that time. Else if I want to play something straight off with the TT cold I use 49.9Hz.
AMR-iFi R&D
|
Post by Owen Y on Aug 7, 2017 20:10:33 GMT 12
colinf - with what I believe is a single phase AC mtr, I presume that mtr speed is not normally variable (aside from the braking adjust) & mtr speed is determined by the AC freq (50Hz). So, an AC freq control would be a nice way to control mtr speed?
|
Post by colinf on Aug 8, 2017 8:47:32 GMT 12
Yes the 301 motor is a shaded pole synchronous design which has its rotational speed determined by the rate of attraction and repulsion of the poles on the armature to the electromagnets. The electromagnets are run on ac 50Hz. So the speed of the motor is largely determined by the frequency to the motor. In a rim drive TT the consistency of speed is determined by the frequency and voltage, both of which are distorted on the mains. Check out the distortion on the mains on an oscilloscope with a 50hz digital generator connected to the timebase trigger input! There is no slack in transmission from motor drive to the platter like there is in a belt drive TT. The belt takes up the flutter variations to an extent. The intermediate wheel in a rim drive TT is much firmer. So having a nice sine-wave of consistent amplitude and frequency becomes important for rim drive TTs. Also, because the platter isn't being asked to accelerate and slow down randomly as it does from a motor that generates flutter from impure mains, this horizontal mechanical flutter isn't being transmitted to the tonearm base on the same plinth. So the tonearm resonance isn't constantly being excited. That allows the cartridge to trace the groove in a more stable fashion as well. It's easy to hear the difference with and without the power supply on my Garrard, the soundstage is clearer and piano especially becomes more solid.
AMR-iFi R&D
|
Post by Owen Y on Aug 8, 2017 9:55:54 GMT 12
Yes, piano pitch & sustain is often revealing. Rim/idler-drive, I found, also preserves 'leading edges' better (than belt drive) - no surprise here.
|
Post by Graham on Aug 8, 2017 10:56:59 GMT 12
Hi Colin. How difficult would it be to design a simplified speed control unit for a synchronous motor as fitted to a Thorens, AR, or Rega ? The Papst DC motor in my Oracle is inclined to hunt very slightly which I feel is a problem with nearly all DC motors no matter how well designed the control unit may be. I like the concept of using a synchronous motor which will lock onto the supplied frequency regardless of voltage fluctuations, however being stuck with the mains frequency is not ideal. I have a brand new Thorens motor which is designed for 115 volts AC using a phasing cap for the 90 degree shift. It is a 16 pole motor that spins at 375 rpm on 50HZ. Is it possible or practical to design a speed control for this type of motor with a suitable wide frequency range to provide 33 and 45 rpm ? Food for thought.
|
Post by colinf on Aug 8, 2017 12:34:03 GMT 12
Yes I've been thinking about making a quadrature version of the speed control for a little while now. It would need the big microprocessor with lots of output pins to operate the two resistor-ladder dacs required. DC motor hunting is a negative feedback issue. In order for the feedback to correct an overspeed for example, the motor has to have generated the overspeed in the first place. So the speed is always being brought back to normal with negative feedback. Same with direct drive TTs. It's arguable that a no-feedback synchronous motor fed constant frequency and sinewave voltage would have less flutter. The normal synchronous motor could be fed a single phase with the cosine phase being fed by a capacitor (as happens in most of these design TTs eg Linn), but when changing from 33 to 45 the cosine phase would require a smaller size capacitor to drive it. Hence a digital quadrature oscillator would be good. I'll keep you posted!
AMR-iFi R&D
|
Pundit
Post by rocl on Aug 8, 2017 15:41:22 GMT 12
depending on cost, i would be very interested in your box. please let me know if you are making a few more.
|
Post by colinf on Aug 8, 2017 18:11:50 GMT 12
Ok I'll work out the costs to build some more...stay tuned.
AMR-iFi R&D
|
Post by Citroen on Aug 8, 2017 19:04:28 GMT 12
I could also be interested, depending on cost.
|
Post by colinf on Aug 8, 2017 19:20:58 GMT 12
Nice, well I just want to revise the circuit board design before I make any more, which I'll get onto shortly. The last 2 versions I made I chose to use tiny surface mount parts which proved difficult to assemble and solder, and also I'll update the strobe board design on the mark 3 version. I'll get onto it shortly. Estimate about $600 or so.
AMR-iFi R&D
|
Post by Owen Y on Aug 8, 2017 20:04:20 GMT 12
So, to recap, we are talking here about a 230vac PSU, with freq & voltage adjustment? Or, a low voltage DC PSU?
|