dam 1021 R2R More Mods…
UPGRADING THE POWER SUPPLY
I Have been pretty happy with my current implementation of the Soekris DAC, using a basic “workhorse” 7812/7912 based power supply. These are legendary for their ruggedness and still the most widely used regulator family for most commercial HiFi equipment.
But not having a lot of time and/or patience to try different mods, I mainly rely on good engineering and the collective wisdom of the people at the diyaudio forums. And lately, improvements in sound have been reported using low noise supplies with the Soekris DAC board [link].
In addition, there is always the attraction for “new and improved” and the introduction of low cost, low-noise supplies in recent years makes upgrading an almost a “must do” thing for diy people like us 🙂
I decided to upgrade the power supply section with a TPS7A47 based power supply kit from diyinhk and an R-core transformer. The TPS regulator has very low noise and a wide operating range. The R-core transformer has multiple (4) secondaries that can feed multiple supplies.
Some users are using shunt regulators for this board. But with the local regulators being of the series type, I don’t think the benefits of a shunt regulator (very low impedance, with “instantaneous” power delivery) can be realized. (Besides I would need to build from bare PCBs and that would take a long time).
The diyinhk kits are based on the factory recommended PCB layout and as shown in the evaluation boards. As such they should perform as specified in the datasheet.
The kits implement some enhancements over the evaluation board such as built-in AC rectification, large filter capacitors, heatsinks, screw connectors and noise-reducing film capacitor. I am using the +/- 12v-15v bipolar supply kit.
The bipolar supply uses TPS7A47 regulators for both the positive rail and the negative rail. I had inquired diyinhk why not use the TPS7A33 [link] for the negative rail and they replied saying that the chip was very frail and had destroyed a few during prototyping. These kits have been available for a while, implying that they in fact work well.
The transformer has 4 secondaries. I can add two more supplies later… These transformers have the following specs:
- Primary : 0-115vx2,
- Secondaries : 0-9X2(0.7A),0-15X2(0.7A)
What is so good about R-Core transformers?
Here is an excellent resource comparing the different types of transformers [link]. I have summarized here the main advantages of the R-Core:
|Leakage flux||The Balanced winding cancels the leakage flux. The total amount of leakage flux is extremely small. (1/10th compared to EI)||The winding is not balanced hence there is lot of leakage flux.||There are gaps in the magnetic path and the winding is not balanced resulting in very high leakage flux. Requiring shielding.|
|Efficiency||Very low losses result in better efficiency. Efficiency > 90%.||Efficiency is better than EI transformer but less than R-core transformer.||More losses result in poor efficiency.|
|Exciting Current||Exciting current is minimal as magnetic paths align with the rolling direction of the steel. Also gapless core||Less than that in EI transformer, but higher than that of R-core due to lack of balancing of winding.||Higher exciting current is required due to presence of magnetic gaps, the inability to use the effect of rolling direction of grain oriented steel, etc.|
|Heat generation||Heat generation is minimal due to low Iron loss. The large surface area of the coil allows for better heat dissipation.||The core is not exposed at any point. Thus the heat generated has no area for dissipation.||Due to iron loss the heat generation is more. Large part of the winding is covered inside the core resulting in poor heat dissipation.|
|Size||40% smaller than EI transformers|
I first tested the supply with two 75 ohm resistors (at 12v this results in 160 mA of current -which is approximately the current consumption of the DAC board). It is always a good idea to test everything while building the circuit. This will prevent frying some component (or even frying yourself). Always exercise caution when dealing with line voltages.
5V Auxiliary supply (for the “dirty” side of the isolators)
I have been using the on-board output buffer connected directly to high-impedance headphones (Senn HD-580, rated at 300 ohm). The output buffer, the LME49724 [link] is specified to drive a load as small as 600 ohms while meeting specifications. At 300 ohm, it should be pretty close to spec.
Using the Philips Fidelio X1 (with a measured impedance of 37 ohm [link]), the direct connection seemed less pleasant (than the Senn headphone). Thus the need for a “real” head amp or at least a buffer with sufficient driving capability.
The diyinhk D1X is an inexpensive head amp implemented with a couple of LME49610 high performance audio buffers.
The kit comes with “basic” dual opamps. I had some OPA627s which I obtained when they first appeared in the market, and used those instead.
Due to the size of the dual chip carriers, some of the capacitors had to be installed on the back side.
I mounted it on the back of the case (because there is already a hole for the plug).
Direct connection to the DAC’s raw outputs:
RUNNING TOO HOT…
After hooking up the power supply to the DAC Board and head-amp, the temperature at the heatsink of the power supply was running up to 75C and this is with the case cover off. Although likely a “safe” operating temperature since the specified maximum junction temperature is 125C, this is way too hot for my liking. In addition, with the cover on, it will be heating everything inside the case which will make run even hotter. Regardless of whether it can run at this temperature at extended periods of time, it is just not good practice running your parts are this high temperature.
The 15V outputs of the transformer measured about 18V AC (This results in about 25V DC). This means that with the TPA, being a low drop out regulator, approximately 12V is applied across the regulator itself. With a 12 V output, the amount of power dissipated by the regulator is just as large as the power dissipated by the DAC board – no wonder it gets so hot.
Actually, this would be a very good transformer for a Salas Shunt implementation. From the manual of the Salas, it specifies:
Voltage drop across regulator: Get a transformer that for DC gives you at least 5V extra from your target voltage. 7-10V is better against mains fluctuations, transformer regulation with high current etc.
But the TPA regulators which is a low drop out regulator only requires about .3 V of headroom and 12 V extra is totally unnecessary.
I therefore used a different transformer, this one has 2X 9V AC secondaries and a primary specified to run at 220V. According to the label, each 9V output can supply 0.85A which is plenty for our application (30VA total power).
I rewired it for 120V operation (I don’t remember exactly what I did, but likely I changed the series input connection into parallel connection). The measured output was almost 12V AC loaded. Perfect.
Now running at 35C… Perfect!
ELIMINATING RINGING IN TRANSFORMER SECONDARIES
The correct way is to use a properly designed snubber, but the second best and much easier to do is to use a small value capacitor across the secondary [link]
The ringing is caused by the parasitics of the transformer + rectifier circuit. The parasitic inductance of the transformer windings and routing set up a resonant circuit with the stray capacitance of the circuit. When the current in this resonant circuit is interrupted, as when the diodes turn off, the energy left in the resonant circuit will cause it to ring until the energy has been dissipated in the parasitic resistance of the circuit. The ringing tends to be in the 100s of kHz to low MHz range – so well outside the audio band.
There are several ways to address this ringing. One can add a small capacitor (100 nF is usually a good starting point) across each secondary winding. This lowers the frequency of the ringing and effectively suppresses any coupling into the amplifier.
100 nF film caps across the secondaries:
I also added a capacitor across the primary:
NO MORE POPS
(At least with headphones)
In addition to several enhancements over the original D1 headamp, the D1X implements an output signal relay:
The new distinctive feature is our customized power on delay and instant off output relay circuit. It can prevent the annoying power on/off clicks and pops, and with the DC servo circuit, expensive headphones are further protected from damage.
Especially for this DAC, this feature comes as a real bonus. The incredibly loud POPs during power on/off are totally eliminated.
MY OTHER POSTS ON THIS DAC
|Soekris dam 1021 R-2R DAC ILLUSTRATED GUIDE||Users Manual||[link]||Users manual for the Soekris DAC.|
|Soekris dam1021 Build||Build Guide||[link]||Details of my initial build of the Soekris DAC.|
|dam1021 R-2R DAC MODs||Mods||[link]||Mods I have performed on the DAC build.|
|dam1021 R2R More Mods||Mods||[link]||Later mods on the DAC build.|
|Digital Filters for Soekris R2R DAC||Digital Filters||[link]||Extensive list of DIY filters from the diyaudio filter brewing forum thread.|
|R2R Benchmark Filters (for now)||Digital Filters||[link]||Latest set of filters developed and shared in the diyaudio filter brewing forum thread. The best filters of the bunch.|
|R-2R DAC For The REST of US||Technical Details||[link]||Introductory post describing the innovations and capabilities implemented in this DAC.|
|The Soekris R-2R DAC: Technical Details||Technical Details||[link]||Additional technical details of the Soekris DAC that were not covered in the post above and collected after I had the DAC on my hands.|