DAC to preamp technical question


Curious as to what is happening in this situation : Using a Musical Fidelity M1 Clic as a DAC (has a volume control as well). Using the fixed rca outputs into a Balanced Audio Technology vk51 preamp into a pair of Hypex Nc400 monoblocks ,into a pair of Triangle celius 202 speakers . The issue is the volume gets loud very quickly and distorts at a lesser volume than i would like . When i use the rca variable volume outs and use the M1's remote at about half volume there is a much greater range on the bat and sounds better . This is my first dac into the newly aquired BAT pre . I have three more dacs coming next week as well . Is the issue the volt output from the fixed outs on the M1 clic ? 

128x128maplegrovemusic
From what i have understood it is the more the relationship between bat and hypex that is the issue ?

No, what matters with respect to the positions of the volume control you would be using are the gains of all of the electronics, the sensitivity of the speakers, and the output level of the source. In this case the gain of the Hypex is about average, and is almost exactly the same as the 26 db gain of the VK-60/75, which is in the ballpark of being average for a power amp. And the 92 db sensitivity of your speakers is just a bit above average. So the major cause of having to use the volume control at low settings is the well above average gain of the preamp, coupled with the use of a digital source having output levels that are typical of digital sources, or perhap a bit greater than typical (and considerably greater than the output levels of most analog sources).

Unrelated to that, though, note in the Stereophile measurements of the VK-60/75 that it has an output impedance in the area of 3.5 to 3.9 ohms, which is very high for a power amp. That will definitely result in a different sonic character in conjunction with your speakers than the Hypex is providing, which like most Class D designs (and most other solid state designs) has a negligibly small output impedance (in the area of 0.001 ohms in this case).  That is all apart from and in addition to differences between the intrinsic sonic characters of the two amps.

I couldn't find an impedance vs. frequency curve for your speakers, just the indication here that "the load appears to be an easy one for any amplifier worthy of the name - the minimum on the impedance curve touches 5 Ohms while the nominal impedance is 8 Ohms." So the tonal differences using the BAT amp with those speakers, relative to using the Hypex amp, will be less than with other speakers whose impedance varies more widely as a function of frequency, especially with respect to other speakers having impedances which dip to particularly low values at some frequencies, but I would still expect the BAT vs. Hypex tonal differences resulting from their output impedance differences to be significant. Again, that is all apart from and in addition to differences between the intrinsic sonic characters of the two amps.

Not being familiar with the speakers, and without an impedance curve, I can't predict the character of those sonic differences, or whether they would be desirable or undesirable. But a home trial with return privileges would certainly seem to be in order.

Good luck. Regards,
-- Al
Thanks for the follow up guys . One last question . I plan on picking up the a BAT 60/75 tube amp . It is the companion amp to the vk5i . When that is put in i assume volume control will increase ? From what i have understood it is the more the relationship between bat and hypex that is the issue ?
Maplegrovemusic, no one is suggesting that the BAT has a design flaw.  Based on your latest posts, it appears that there are two separate issues:

1)A defect in your particular unit, which seems likely to account for the distortion that I had indicated earlier I was unable to explain.

2)Gain that is higher than desirable when used in conjunction with most digital sources, given also that amplifier and speaker sensitivities are average or higher.  I would expect, in the absence of defects in your particular unit, that the BAT could "handle" maximum inputs of 2 volts or more, but the consequence of its high gain would simply be having to operate the volume control in the lower part of its range.  That is a very common problem these days, when digital sources are used, and some well regarded preamps (such as many Conrad Johnson models) have gains that are even a good deal higher than the gain of your BAT.

That is the main reason that Rothwell attenuators and other similar devices have appeared on the market.  About which, BTW, reported experiences seem to be mixed.  I and a number of others here have used them with excellent results, while some have reported that they compromise dynamics.  I suspect that the very high input impedance of your BAT would make it more likely than not that they would provide good results in your case.

Regarding the NC400, the last few pages of the datasheet include descriptions of how the surrounding amplifier chassis can be designed so as to provide unbalanced RCA inputs, although the datasheet discourages those approaches as being non-optimal in terms of performance.  If your monoblocks provide RCA connectors, presumably/hopefully they use one of those approaches.  If they provide only XLR connectors, the only safe way that occurs to me of connecting a component having unbalanced outputs to them would be via Jensen transformers, or equivalent, such as a pair of the Jensen model PI-RX, similar to PI-XX shown here but with an RCA input connector, as depicted in the lower right corner of this datasheet.

And in fact if the DAC you end up choosing includes a volume control which seems to provide satisfactory performance, but it only provides unbalanced outputs, you may want to consider using a pair of those transformers and doing away with the BAT altogether. 

Regards,
-- Al
  

Just saying it’s bad news not to be able to use RCA interconnects with it, which is fact, looking at what Al has rightly pointed out.

What I think about the sound of Class D has not even been mentioned here, if you want to bring it up there are other threads to do it in. 

Cheers George

 

I am a little confused with the determination of the BAT pre having a serious design flaw . Should it's cd input not be able to handle a standard 2 volts ? Or am I misunderstanding what I read here ? if so what would be a good match volt wise ? Also i am using the supplied balance to rca jacks that BAT recommends and claims has no derogatory influence on the sound . I do have three dacs coming later in the week . Two of them have  balanced outs , so I will be giving those a try as well .
Thanks for the responses fellas . A new twist in my journey . I had just purchased the BAT used from a hifi dealer in town last week . I inquired if they knew the tube life on the tubes . They did not know . So yesterday after about an hour of listening the left channel just went dead and gone . No warning signs what so ever .Shut her down , left and came back an hour later and started up again . This time it only took ten minutes for the same thing to happen again . I am assuming the tubes are in need of replacement . Might have some correlation as to why the BAT sounded worse with the higher sensitivity speakers ? Goerge , you really need to get over your obsession with putting down Hypex . LOL

"but after looking at section 7.3 of the NC400 datasheet "


Yeah, looks like your right Al, what a pain, not being able to connect normal RCA’s, as it’s ground floats at this point and this is where the shield (ground potential) on external equipment has to connect.

I don’t understand why they would design it that way, how would they sink any RF to ground when using XLR anyway, suppose they had their reasons, not good for sales though in my opinion.


Cheers George

I found this when reading the specs of the M1:

Fixed outputs, not controlled by volume at 0dB (same source as main out)

Not sure what that means?
After looking at both the website description of the M1CliC and the manual, apparently what this means is that if an analog input is selected (the unit provides both analog and digital inputs), the level of the fixed output will be the same as the level of that input.  Unfortunately, though, the level of the fixed output when a digital input is selected is not specified.

In any event, after looking at the relevant specs for all of the components that were mentioned, consistent with George's comments it appears to me that the root cause of the problem is the high gain of the BAT preamp (18.34 db per John Atkinson's measurements of the SE version, as published in Stereophile), coupled with the fact that nearly all digital sources having non-adjustable unbalanced fixed outputs provide "full scale" (maximum) output levels of at least 2 volts, and often more.

That would certainly explain the limited range of the volume control that can be used when the fixed outputs are boosted by 18.34 db and then applied to a 200W/8 ohm amp having 25.8 db of gain powering the 92 db/1W/1m/8 ohm Triangle speakers.  I'm not sure, though, how to explain the distortion that was mentioned, which occurs with the Triangle speakers but not with the other lower efficiency speakers.

Also, in most cases I would second George's suggestion of trying the system with the preamp removed altogether, but after looking at section 7.3 of the NC400 datasheet I'm not certain that would work, given that the M1CliC provides only unbalanced outputs, and depending on how the power amp is configured externally to the NC400 module.  I'll mention also, BTW, that the BAT's output impedance rise he cited is not an issue in this case, as the NC400 has very high input impedance. 

Finally, it appears that the BAT only provides balanced inputs, while as I mentioned the M1CliC only provides unbalanced outputs.  Which leads me to wonder how well the BAT may be performing when provided with unbalanced inputs.  If you are using adapters or adapter cables to connect the two, you may want to consider using a Jensen transformer instead (~$275), which would provide the BAT with a true balanced pair of signals.  To reduce the signal level, you might then also want to try a pair of Rothwell attenuators, in their balanced version, inserted into the input jacks of the BAT. 

Or, alternatively, just use the M1's variable outputs into the BAT, as you have done, and be happy.  :-)

Regards,
-- Al


Your better of driving your Hypex Nc400 direct from the variable output of the M1. As it’s output stage has better low output impedance drivability than the Bat has, you’ll also loose some of that excess gain doing so as well, and gain more transparency.

Musical Fidelity M1 Measured:                                                                                                                           The impedance from the variable jacks was 47 ohms at high and middle frequencies, rising to 76 ohms at 20Hz.

Bat VK51 Measured Test:                                                                                                                                Output impedance was twice that specified, at 410 ohms over most of the audioband. This is low, though it did rise to a high 4.7k ohms at 20Hz, due to the size of the output coupling capacitors.This will not be an issue with components having high input impedances, such as BAT’s own power amplifiers, but it does mean prematurely rolled-off low frequencies with low load impedances.


Cheers George




  

Want to add that using the fixed outs of the M1clic with less sensitive speakers sounds great . King Sound electrostats and Magnepam MMG . Using both speakers the sweet spot for the volume is around 50 out of the 100 on the digital display for the BAT . Can go louder as well without distortion . So in a nutshell is the high sensitivity speakers the cause ? and is there a solution besides using a different pre ? Possibly different amps with the triangles ?
I found this when reading the specs of the M1:

Fixed outputs, not controlled by volume at 0dB (same source as main out)

Not sure what that means?  Maybe Al, (Almarg) or someone with more tech knowledge than me can interpret this.  But, I would assume the fixed output has way too high of a gain for the BAT.