... an odyssey with spectral dmc10

[Jon Q]

After purchasing a spectral DMA 150 (about 18 months) and made an upgrade, I then decided to look for a preamplifier. Since my adolescence this manufacturer has always inspired me. So the preamplifier had to be also ... spectral.

I had access to 3 models for my final decision. DMC-20, DMC-30SL and DMC10 serie Delta. The DMC30 is very transparent and fast, but it doesn't have the (neutrality I'm looking for). I also liked the DMC20 and DMC10, both have an identical signature, which I look for.

With an update it is possible to improve your performance on these two models. My final choice was the DMC10 Delta series, the reason was that the DMC20 the phono board was not included.

After a long google search, I finally found a DMC10 for sale in Denmark. The owner informed me that he had some problems and two / three risks on the front panel. As I was going to do an update, I risked the purchase.

Being an equipment from the 1980s, the electrolytic capacitors were destined to be replaced, there were other components to be replaced, RCA connectors, resistors, dip-SW.
Two days after it was delivered, I decided to do a first audition. In fact, he has some problems that the seller mentioned, including leaking transistors. This is a problem, because they are models that are no longer on the market for sale.

j

 

[ack]

Are there equivalent?

 

[Jon Q]

Are there equivalent?

yes, there are alternatives...

 

[Jon Q]

The Spectral DMC10 , uses Vishay RN55D resistors (1% - 100ppm).
The line amplifier the resistors (RN55D) were replaced by the RN55E version, those that work in current mode, and those that work in voltage mode, were replaced by RC55Y series of welwyn (0.1% - 15ppm), signal and feedback path.

In the phono stage, the upgrade was the same, except for resistors correction- RIAA, vishay VTA series , (pic. 6)

http://www.vishaypg.com/docs/63011/vta5256.pdf

The resistances of the main board, (circuit protection) kept everything the same.
The signal input adjustment pots have been replaced by resistors rc55y with the identical values of the DMC30.

Another situation was the replacement of the gain pot in the line amplifier, by a resistor from welwyn rc55y, 0.1%. (pic. 7)
Line gain is preset at 25dB
The polyester capacitors were replaced by polypoprylene capacitors vishay, 1837 series (pic. 8) and .... also a shower for the mechanical parts.

j

 

[ack]

Very nice! I have to admit I am smitten by Vishay resistors and power diodes. Incredible clarity and speed.

 

[ack]

Are you using Fluke multi-meters? Which ones?

 

[ack]

From the datasheet, I see these RN55D, like others in their lineup, "are provided with high purity copper leads" and are non-inductive. Spectral has been light years ahead...

 

[Jon Q]

Are you using Fluke multi-meters? Which ones?

Yes, I have 3 Fluke, 77; 87; and 87III. They never gave me any problems.

To analyze other parameters I use a Tektronix TDS5054B-NV AV. It is my tool in digital domain, when I upgraded to my WADIA 6 just as a CD transport, the DAC board was removed, the output driver (AES dig. Out) was designed for this wadia, new clock, three new psu low noise, upgrade of the reset circuit, .....

j

 

[Jon Q]

From the datasheet, I see these RN55D, like others in their lineup, "are provided with high purity copper leads" and are non-inductive. Spectral has been light years ahead...

yes you are right, but when we replace it with the RN55E version the result is surprising, even with the RN55C series it is always superior with the 55D. The RN55E 0.1% "has only two disadvantages".
It is not easy to find all the values on the market, and the other disadvantage is their price. About 10x ... 15x more expensive.

j

 

[Jon Q]

To improve the performance of the phono stage, the dip-SW was removed (fig. 24), it is now a gold-plated jumper.
I am not a fan of dip-SW for the selection of cartridge loads, because they always cause problems with time, even worse when using MC cartridges.
The 2x module NPD5566, socket dip 8 was removed, and directly soldered (pic. 25). The advantage is less contact, and less signal loss. The load resistors were replaced by the RC55Y 0.1%, very very low noise, from Welwyn.
The selected values were .... 51.1R; 105R; 390R and 800R. (pic. 26)

For regulation, the regulators kept the same reference, but from another manufacturer (Linear Technology), the diodes were also replaced, RHRP860 ultra fast from ON. About the resistances of 400R, were canceled. (pic.27).

to be continued.....
j

 

[Jon Q]

The dmc10, each output stage, uses 3 mosfet, 1x VN1304N2 and 2x VP1304N2.
The two VP1304N2 mosfets are in parallel, to reduce the (R / ds "on" 25ohm). Some of these mosfets, having problems, were replaced by VN0109 and VP0109 with the TO92 package, (pic. 38).
The spectral dmc20 also uses these mosfets but in the TO220 package (pic.39). Also the resistors values on the pins, gate, drain and source of these mosfets have been changed, very similar to the DMC20.
Unfortunately, there were also 2 double MP350 transistors, with problems, the replacement was 2N3810, (pic.30).

j

 

[Jon Q]

When replacing electrolytic capacitors, two nichicon models were tested. First for the FG series for 2 months for a testing period, then another series (FW and KA). At the end of another 2 months of testing, the choice was nichicon (FW and KA).

j

 

[Jon Q]

The rca connectors have been replaced by rca whit teflon insulation, the power connector has been replaced with neutrik 5 pins.
Another feature of the dmc10, it has two line outputs. The ac coupled output, the connector RCA, has been replaced by XRL in dc coupled mode. Here the impedance of pin 3 is the same as pin 2. Later on I can inform you about this detail.

j

 

[ack]

Impressive again

 

[Jon Q]

To improve the design on the two leds, a layout was created based on a model already made.

 

[Jon Q]

Another situation was to isolate the transformer, from the chassis. (fig.15_2).
The xlr connector was replaced by a neutrik, the IEC was replaced by oyaide connector, gold / paladium contacts, a 4 mm connector was also added to the ground. (Preamplifier-psu) (fig.17)

 

[Jon Q]

The original cable was replaced by neutrik connectors. Later a new cable (up-grade) was made with kimber wire, Kwik 16 (only the 4 internal wires were used), an insulated wire was also added for the ground connection. (Neotech STDCT-16 with connector mundorf-banana 4mm gold). (fig.22, 23)

 

[Jon Q]

The original impedance of output 100 ohm is now 51 ohm, so also the output XRL pin 3 is added a resistor with the same value in connection with pin 1, (ground).

 

[Jon Q]

On the monitor output the TL071 was replaced (fig.28_1). Here the choice was based on several samples. In the end, the choice was between OPA627BP, personally I don't like the OPA627AP version. and another model LT1122. This model has several versions, interestingly the balanced input module of the spectral DMC30 (except the version DMC30SV), uses this model LT1122BCN8.

There is also an improved version, LT1122ACN8. At the end of 2 months !!! in testing, my choice was for LT1122ACN8. This I.C., has the possibility of increases slew rate and bandwidth by about 25%, adding a resistance between pin 7 and 8 (my choice was 301 ohm rc55y), with a surprising result. (fig.29).
Possibly a good upgrade to the balanced module of the spectral dmc30.

Also the polyester decoupling capacitors (fig.28_1), for TL071 (now LT1122ACN8) were replaced by mica capacitors (fig.29)

j

 

[ack]

Such incredible workmanship! Congrats