Tuner Heads,
Has anyone converted an Acuphase T-1000 from 76-90 to 88-108 MHz?
The service manual details the front panel button combination PLUS the Accuphase part numbers of 15 caps, coils and a resistor that are different in Japan than Europe or North America.
Right now I'm running a Japanese T-1000 at 89.7 (The Classical Station) with an extra 25 usec deemphasis, and it sounds better than any other tuner I've ever heard. I switch to another tuner for the other stations.
Has anyone who is good at this converted one? I don't want to try this unless someone does it and documents it here.
Paul
Has anyone converted an Acuphase T-1000 from 76-90 to 88-108 MHz?
The service manual details the front panel button combination PLUS the Accuphase part numbers of 15 caps, coils and a resistor that are different in Japan than Europe or North America.
Right now I'm running a Japanese T-1000 at 89.7 (The Classical Station) with an extra 25 usec deemphasis, and it sounds better than any other tuner I've ever heard. I switch to another tuner for the other stations.
Has anyone who is good at this converted one? I don't want to try this unless someone does it and documents it here.
Paul
Difficulties / Disadvantages:
1) The front-end (sealed metal box) will have to be opened and tweaked to adjust the antenna filter for the new range.
2) The local oscillator of the superhet mixer will also require adjustment to fit the new range. Note that this is a DDS oscillator and NOT a tank circuit.
3) The MHz value display on the front panel may no longer correspond to the frequency of the actual station.
1) The front-end (sealed metal box) will have to be opened and tweaked to adjust the antenna filter for the new range.
2) The local oscillator of the superhet mixer will also require adjustment to fit the new range. Note that this is a DDS oscillator and NOT a tank circuit.
3) The MHz value display on the front panel may no longer correspond to the frequency of the actual station.
Yes, the front panel button combination changes the readout from 76-90 to 88-108 MHz, but on my tuner it then finds no stations. I'm guessing matters 2 and 3 are taken care of by those 15 components that would be changed.
Those components are the same for North America and Europe except for 1 of them, which may be for the different deemphasis. The special button combo press might work for Europe to North America except maybe the deemphasis may stay at 50 usec instead of 75 usec which would make it a bit bright.
Those components are the same for North America and Europe except for 1 of them, which may be for the different deemphasis. The special button combo press might work for Europe to North America except maybe the deemphasis may stay at 50 usec instead of 75 usec which would make it a bit bright.
What exactly do you mean by "an extra 25 usec deemphasis"? Have you changed the deemphasis time constant of the tuner to 75 us, or passed its output signal through a circuit with a pole and with a zero that covers the original deemphasis pole, or put a 25 us first-order low-pass after the radio? This last solution can't work; a cascade of two first-order filters is something else than a single filter with a larger time constant.
No, I have an RC filter in the output cable plugs! It always has an active crossover and an ambience processor in parallel for 16K ohm load. The first pole(s) are way upstream in the tuner, so it's not two or three in a row. I can't find the scratch paper with the values - the series resistor is 1.6 K ohm and the parallel cap is 0.015 uF, I think. It went from a little bright to just right.
Still, I would rather replace those 15 components, copying an expert who has done this, and get the flat response that way.
Still, I would rather replace those 15 components, copying an expert who has done this, and get the flat response that way.
I understand that you don't have two RC filters connected to each other without any buffering, but that is not the point. When the tuner has a corner frequency of 3183 Hz (50 us de-emphasis) and the extra filter has a corner frequency of 6366 Hz (25 us de-emphasis), you get a different response than with just a tuner with a corner frequency of 2122 Hz (75 us de-emphasis). First-order roll-off from 3183 Hz switching to second order from 6366 Hz simply isn't the same as first-order roll-off from 2122 Hz.
Using two resistors and a capacitor, you could make a filter that does change the overall response into a first-order roll-off from 2122 Hz. It would have to be something like this:
The 3.3 kohm resistor should theoretically be 3.333333... kohm, so you get a 50 us time constant with the capacitor. The tuner's output impedance plus the 1.8 kohm, in parallel with the 16 kohm load, in series with the 3.3 kohm should theoretically be 5 kohm, resulting in a 75 us time constant with the capacitor. It's spot on when the tuner output impedance is around 100 ohm.
Last edited:
If the button combination really takes you from Japan to North America then you should be receiving at least the 89.7 that you've been listening to in the Japan mode. If not, something is wrong or something else is required in addition to the combination.
Marcel, OK, I have 2 slopes and it's not ending up at 75 µsec. I'll cut off the cable plug with the RC filter and make a new one with these components.
1) Is that capacitor 15 nano Farads, same as 0.015 µF?
2) The T-1000 unbalanced output impedance is 200 ohms. Should I get a smaller resistor than the 1.8K ohm, and if so, what value?
Thanks!
Paul
1) Is that capacitor 15 nano Farads, same as 0.015 µF?
2) The T-1000 unbalanced output impedance is 200 ohms. Should I get a smaller resistor than the 1.8K ohm, and if so, what value?
Thanks!
Paul
MarcelvdG, I can't help asking, I have a Topping E30 II DAC and would like to experiment with that and other DACs with the digital output on the T-1000. The deemphasis, whether 75 or 50 µsec, is already there in the digital out, I assume.
1) Could I do the same thing with the Topping DAC outputs as you diagrammed?
2) Output impedance of the Topping DAC is 20 ohms. Would that maybe 4% change in source impedance be a significant change in frequency response?
Appreciate You,
Paul
1) Could I do the same thing with the Topping DAC outputs as you diagrammed?
2) Output impedance of the Topping DAC is 20 ohms. Would that maybe 4% change in source impedance be a significant change in frequency response?
Appreciate You,
Paul
Yes.
I don't know if you have resistors with E96 values at your disposal or prefer E24, so I just calculated the optimal E96 and E24 values. As usual, the E96 values are closer to the ideal.
Ideal values with 200 ohm source and 16 kohm load: 3.33333... kohm and 1.660465116 kohm
Best E96 values: 3.32 kohm and 1.69 kohm
Best E24 values: 3.3 kohm and either 1.8 kohm or 1.6 kohm
(That is, when you account for the effect of rounding off 3.333... kohm to 3.3 kohm, the ideal value of the other resistor changes to 1.702097902 kohm, which is practically in the middle between 1.6 kohm and 1.8 kohm.)
The capacitor has to be an accurate film capacitor or an accurate class 1 ceramic capacitor (such as NP0, also known as C0G), definitely not a class 2 ceramic capacitor (which have dielectric codes such as X7R and X5R).
Last edited:
Of course.
If you would connect a filter that works perfectly with 200 ohm source and 16 kohm load to a 20 ohm source and 16 kohm load, you would get 0.2573 dB too much treble.
Coming back to the E24 resistor values: when the source impedance can be either 200 ohm or 20 ohm, 1.8 kohm is to be preferred over 1.6 kohm.
Last edited:
Well that depends on whether there're stations in that band. You could use your cellphone to find the stations and frequencies in your area (since FM is local).
However, if 89.7 is not receiving then the button combination clearly applies only to the display and not the DDS local oscillator. The antenna BPF may also need adjustment which is a problem as changing selected components could compromise the impedance matching between stages.
There're some cheap tricks with adjusting filters but your unit seems to be an expensive one.
Not to sound pessimistic, but there's one more catch, the MCU firmware that runs the DDS. This code not being serviceable, may not even be mentioned in the service manual.
Adjusting de-emphasis is OK though.
Why would they use different firmware for different parts of the world? That only complicates matters, also for Accuphase.
I don't know Marcel, but as I understand there're only few ways of changing a DDS' output frequency and it's almost always done programmatically, seldom using passive resistors / capacitors. For example,
1) Write a new word into its frequency / PLL calibration register (or alternatively shorten the staircase counter's modulo value).
2) Switch the DDS clock to a new scaled frequency value.
I agree with that it's inconvenient to have different firmware / versions for the same model but sometimes you'd have to, even if the changes are minimal.
1) Write a new word into its frequency / PLL calibration register (or alternatively shorten the staircase counter's modulo value).
2) Switch the DDS clock to a new scaled frequency value.
I agree with that it's inconvenient to have different firmware / versions for the same model but sometimes you'd have to, even if the changes are minimal.
Last edited:
You could check this by opening up the unit and following the signals from the DDS chip. The one shown in the datasheet is the AD9850 which has current outputs on pins 20 and 21, that would have to go through an I-V converter (usually an op-amp) somewhere nearby.
From the frequency of this signal you might be able to assess the LO and its frequency steps. You could also switch to the Japan (working mode) to see how things change.
Thanks for the post. I was under impression that the conversion is done by a combination of front panel buttons maneuvers and only now I see in the manual that they use different boards for different regions. I agree it's a nice tuner . It's biggest fault is that one can't escape the desire to match it to thier Class A integrated and of course grab a useless SACD player as well selling a kidney . The main reason I'm selling mine. I should have kept T-109V . Much prettier and matching amps are half the price .
I don't think you need to. You can just write a program that covers the DDS settings for all FM bands and inform the microcontroller what FM band is in use. A few bits will suffice for that. What I don't know is why the Japanese setting can't receive the 89.7 MHz, though.
Yes, this frequency is common to both bands and should work anyway. Even if we assume that the antenna filter is still in Japan, it ought to pass 89.7 through isn't it ? This is why I assume the LO could be wrong and doesn't result in an IF signal within the expected range, possibly causing the receiver mute the audio.
Probing the DDS is not a big deal and could be done in a non-invasive way.
Maybe the pricing is significantly different between regions and Accuphase wants to prevent customers from converting from one region to the other ? Money always seems to make perfect sense whenever / wherever nothing else does !!