Burson Audio Opamp module and buffers

Discrete opamps - are they a step forward?

Product: Hyper Dynamic Amplification Module and buffer
Manufacturer: Burson Audio - Australia
Cost: $145.00AUD (107 USD) + postage (for a pair of dual modules)
..........$285.00AUD (210 USD)+ postage (for the buffer in kit form)
..........$335.00AUD (247 USD)+ postage (for the completed buffer)
Reviewer: Nick Whetstone - TNT UK
Reviewed: May-July, 2006

Modifying CD players to achieve a higher performance is now quite popular amongst the DIY hi-fi community. Most budget players can be improved significantly by upgrading the power supplies and there are a host of other components that can be replaced to good effect. One of the most popular upgrades involves replacing the original opamps with something better, and a whole cult has grown around the subject of 'which opamp is best'! Some people argue that all opamps have their limitations and cannot compete with a discrete circuit. I have known cases where the opamps were removed and replaced with a discrete circuit but this is only possible if you are confident of building a discrete circuit that out-performs the opamp. And of course, there must be enough space inside the CDP or DAC to install it!

Now, wouldn't it be great if we had a discrete circuit that took up little more space than the opamp that it is replacing? Well, not only did the guys at Burson Audio think that, they came up with a working example! And that's what the Hyper Dynamic Amplification Module is. Please note that this opamp module is nothing to do with the HDAM technology used in Marantz products.

Having established that the opamps in my modified Philips CD723 were dual types, Burson Audio kindly sent me some samples to try out. In theory at least, it's quite straight forward to install the opamp modules. You simply desolder the existing opamps, install DIL sockets in their place, plug the opamp modules into the sockets and solder the ground wire to a suitable grounding point, eg the output phono socket. In practice of course, things are never quite as simple. It had been a while since I had the lid of my CDP off and when I removed it, it was clear that I didn't have enough space to fit even the opamp module's with their compact footprint. (I should explain that my modified CD723 has also been converted to a top-loader (as described here) and that has resulted in the main PCB being situated under the transport) Fortunately, I had another unmodified CD723 with plenty of empty space so I installed the opamp modules in that.

Now is the bit where I would like to tell you just how much improvement the opamp modules made over the original NJM4560D opamps. But when I powered up the Philips and pressed play, all I got was a slight hum with a much reduced output of the music signal. I tried another HDAM but with exactly the same results.

I contacted Burson Audio who were also puzzled by my results but asked if it would be possible to try a different CDP. A friend of mine has a much modified Arcam Alpha 5 and he agreed to try the opamp modules. The Arcam uses single opamps rather than the dual type in my Philips but we managed to separate the modules. There was one problem though. The wiring diagram on the Burson Audio web site was wrong and when we powered up, there was some smoke and a regulator went to the great PCB in the sky. We contacted Burson Audio and I believe that the mistake is now corrected. In any case, most people will not need to convert the modules from duals to singles. Having worked out the correct wiring, we converted another module and tried that.

The Arcam Alpha 5 uses four single opamps. Two for the I/V stage and two for the output stage. Once again, we found that when used in the I/V stage, the results were disappointing.

So we moved the opamp modules to the output stage and at last we could hear why Burson Audio are so proud of these modules! There is quite clearly an improvement over the opamps (and we tried quite a few). In (and out) went the OPA132, OPA134, OP-27GN, OP27E, OPA606KP, LT1028, OPA627BP, and the good 'old' NE5534. In truth, only the OP27's gave the opamp module anywhere near a real challenge. And please note, the OP27's sounded very good in the Arcam but I am not saying that they are the 'best' opamp for all CD players!

So what did we hear with the opamp modules? They are very detailed especially in the highs, the sound is more 'effortless', flowing from the speakers without strain. Cymbals sound more real with no splash, and each element, be it vocals or instrument, is more clearly defined in the sound stage. Vocals are clear, slightly more forward than the OP27's. The OPA627 vocals are much darker and flat compared to the opamp modules. Piano (always an excellent test) has that decay that is 'not quite there' with most of the other op-amps.

The most noticeable improvement is the 'blackness' of the background. Power and control are also better which we assumed could be down to the lower output impedance of the opamp modules.

So pleased were we by the results of fitting the opamp modules to the output stages of the Arcam, my friend was keen to see if he could squeeze any more out of it. Attention was paid to tweaking the power supply, particularly in the choice of bypass capacitors. Some polypropylene bypass caps (used on the opamp supply lines) were replaced with military-grade tantalums (seen in the picture) and this brought about another clear improvement. It was as if the opamp module's not only sounded good in their own right, but they allowed other changes to be more audible too.

Taking things a stage further, ALW super-regulators were fitted to power the output stage with the opamp modules and once again a clear improvement was detected. This is good news as Burson Audio are developing their own super-regulators which I hope to review later.

After these modifications, the difference between the opamp modules and the OP27's was even more apparent. So keep in mind that although the opamp modules are relatively straight forward to install in your CDP, and will almost certainly make a big improvement, you will probably get the best results with a little more 'tuning' as well. This shouldn't be a problem as anybody fitting the opamp modules is likely to have the knowledge and facilities to change other components too! The opamp modules represent a not inconsiderable investment and although they do sound good as they come, it makes sense to squeeze the best from them!

It's worth noting that the opamp module modules seem to improve further when they are well warmed up. And like a lot of other electronics, they seem to benefit from some burning in time.

Having got the opamp modules working so well in the output stages, we were still keen to try and get them to work in the I/V stage too. The guys at Burson Audio worked hard to try and solve the problem and we made several changes to the opamp modules but still couldn't get them to produce better results than the LM6181 opamps, that when used with the opamp modules in the output stages, produced the over all best results. Burson Audio are, at the time of writing, working hard to design a new opamp module specifically for the I/V stage and I look forward to reviewing that when it becomes available.

Buffers

OK, if you are sitting there thinking that this is all great fun for those that don't mind poking around inside their CD players with a hot soldering iron but it's not for you, there is good news! Burson Audio have considered the non-DIYers too and produced a stand-alone version of the opamp modules.

Housed in a smart black box, sporting some very classy phono sockets, and with the almost obligatory blue LED on the front, this unit is designed to go between the CD player (or DAC if you use a separate one) and the pre-amp or integrated power amp. The box is 145 mm wide by 210 mm deep so it won't take up a whole shelf on your hi-fi rack. If you don't have a spare shelf, then it is compact enough to be sited somewhere behind. All you need to do is plug the mains cable in and connect the unit to the CDP or DAC via the phono sockets and also to the pre-amp or integrated amp. And that's it, no soldering, no housing to be built, nothing to do except listen. How refreshing that was for this reviewer!

I first connected the buffer between a Monica2 NOS DAC and a home-brew pre-amp. Even though the opamp modules had clearly shown that they can provide a significant improvement inside the CD player, I was wondering about adding yet another circuit into a system, together with the extra interconnects that they require. But I needn't have worried. From the first track that I listened to I could clearly hear the improvement.

I spent an hour building a simple switching unit so that I could switch the buffer in and out but I very quickly tired of playing with that. Basically, it was no contest and the buffer stayed in circuit for the rest of the day. If you do want to AB test the buffers in your system, do be aware that they have a small amount of gain (3 db) and take that into consideration.

The buffer provides many of the improvements that can be heard with the modules. Detail is improved. I don't know how many times I will say this, but once again I was hearing detail in very familiar test recordings that I hadn't noticed before. There was also a sense of being closer to the music, both physically and emotionally. Timing was improved and the music just seemed more 'right'. And yes, there was more bass! In fact the music just seemed to flow more freely from the speakers. It was one of those occasions when instead of flicking through some favourite test tracks, I listened to the whole track. The sound stage was larger, this was easily checked using the AB switching unit. It was also deeper.

I repeated the experiment with the buffer between an un-modified Philips CD723 and the pre-amp. The results were the same. There really wasn't any need for the switching device, the improvements were easily discernable. Then I took the advice of Burson Audio to remove the pre-amp and replace it with a passive volume control. In went a 10K stepped attenuator and the improvement was even greater. Without another active circuit in the system, clarity was slightly better and most amazingly, I got the impression that there was even more bass. To get more bass by replacing an active stage with a passive volume control, seems to defy science but that's what I was hearing!

One of my test CD's has two particularly well recorded pieces of orchestral music. The buffers made it even easier to pick out each section of the orchestra. Violins sounded 'separate' instead of one mass, and it was almost like watching the fingers of the lead violinist. When the kettle drum was struck, the impact was clearly heard followed by the sound of the drum, including the decay. The woodwind section also sounded like separate instruments rather than a single sound. And percussion was crisply portrayed against the sound of the rest of the orchestra.

Having tested the buffer in my second system, it was time to try it in the main system. In it went between the NOS DAC and the discrete Jfet pre-amp. This time I was even more amazed. With this system producing bass flat down to 20 Hz, the improvement in bass was even more apparent. It was as though I had been given an extra octave. Bass quality is also improved, as is the whole frequency range. To use an analogy, it's like a pipe that the signal travels through has been enlarged and cleared of any internal debris so the music comes out more freely and less polluted!

Now given that the buffers are claimed to cure the shortcomings of inferior output stages, it is no surprise that it works very well with the cheapo Philips CD723 and the NOS DACs but I wanted to know how it would perform in a more up-market system. So I took the opamp module to a friends house where we connected it between his Musical Fidelity A3.2 CDP and Gamut C2 pre-amp. The A3.2 retails for around 1000 UKP and the Gamut I believe cost double that. Amplification is provided by four Gainclone monoblocks and speakers are the excellent Monitor Audio GR60's. My friend spent a long time putting this system together and it does sound very good so could the buffers make any improvement? The answer was yes although the improvement was much less than with the lower budget components. But given that the CDP and pre-amp already had a very good synergy, it was a creditable performance by the buffer to make any difference!

So why is the buffer so good? Well, first and foremost, it has been designed and built with no compromises by a group of hi-fi enthusiasts, rather than a company who want to build something to sell for a profit. The attention to detail is usually only seen on extremely expensive commercial gear, or equipment modified by DIYers where cost (at least as regards time) is not an issue. Features like a dual mono power supply, carefully chosen audio-grade components (including matched transistors), high-quality PCBs and a filtered power supply ensure that the circuits are operating at their very best.

But there is more going on with the buffers than just a box of quality components. A large part of the improvement comes from reducing the output impedance of the CD player to between 15 and 30 ohms, the sort of low figure usually only achieved by expensive CDPs with a top-quality output stage. The Musical Fidelity X10-D was designed to go between the CDP and amplifier for the same purpose, and I remember conflicting opinions about it's performance, depending on which CDP it was actually used with. You can read a comparison of the X10-D and the buffers on Burson Audio's buffer page.

One example of a system that would benefit from the inclusion of the buffer would be where a NOS DAC, eg Monica2 from DiyParadise is paired with a class-T amp, eg Charlize, from the same supplier. Now the NOS DAC needs to work in to no less than 20K so that's the minimum impedance of the volume control. But the Charlize is best with no more than 10K impedance on the output of the previous stage (the attenuator). The solution is some kind of buffer, or active pre-amp. But with the buffer used after the NOS DAC, the output impedance is reduced to 15-30 ohms, which will work easily into a 10K volume control, which in turn will work well into the class-T amp. I've seen a number of people question the bass output of the smaller class-T amps but I have not found them lacking when set up properly. No amount of input capacitor swapping is going to improve the situation if the impedances are not matched properly!

The buffers are available either as modules, or as a finished, ready-to-play box. The modules require a power supply, casing and connectors but you also then have the opportunity to add a volume control of your choice (and selector switch too if you use more than one source). Burson Audio appreciate that there will be those who wish to take the DIY approach, and others who want a 'plug it in and play' option, so have sensibly catered for both. Summing up, I have been totally impressed by this technology. The modules clearly bettered any opamp tried in the output stage of the modified Arcam Alpha 5. Remember that they work well in the output stage but not in the I/V stage although Burson Audio hope to rectify that issue soon.

If you don't fancy fitting the modules inside your CDP (or perhaps your CDP doesn't have an output stage using opamps), and just want something that you can plug in and hear a big improvement, the buffers are probably for you. Keep in mind that if you own a budget or mid-range CDP, or a NOS DAC, the improvement will be substantial, but less so if you already have a high-end CDP with a top quality output stage. Burson Audio are very sensibly offering a 7 day money-back offer on the buffers so you can try them with minimum risk. My bet is that they won't be getting anybody asking for their money back! My advice is; get this technology before there is a waiting list!