UCD180 questions

Jump to Latest
#2,301
Javin5 said:
On the UcD180HG, I believe the input coupling caps are the two red-brown 0.68 uF caps near the input connector. No guarantee, though.

If you bridge these caps with a piece of wire (dc-coupling), you may have to readjust the offset via the small Burns pot. I intend to do the dc-coupling at some time, but have not found time yet.

Another question to the UcD180HG owners: I found bass response to be quite weak, certainly a lot weaker than with any of my transistor amps and even weaker than with the older UcD180. And yes, I double checked the polarity of the speaker connections and input connections; everything is ok and in phase. What is your experience?
Click to expand...

Which "Burns" pot would that be? There's two of them, and if you get the wrong one it could "burn" more than even shorting those silver caps out would have. It's almost deadly to give such a response without drawing numbered pictures here, sorry to say.

The odds of it being those caps though are far greater than the others mentioned. Now imagine how very little it would take to trace the input pins back to the caps and remove all doubt. I'd worry more about the level of input offset before DC coupling then I would about having to re adjust output offset post coupling.

The level of offset should change hardly at all by removing the caps from the picture, and in my view you should never attempt to use this adjustment to cancel the output offset of your pre amp either. Just think your wires could get swapped eventually.. and I think it better having the comparator as balanced as possible with inputs shorted.

As per your question I've never been so blessed as to be subjected to a solide state amp that had accurate, believable bass. Bloated, tightened, compressed, sluggish, sure, but believable? Nothing close. Said to say it serves little to no purpose making such a statement, it's apples and crackers.

Imagine buying a high performance crate engine, where all you have to do is "bolt it in". You needn't know a damn thing about all the blueprinting or decades of R&D that's gone into it. How many people are going to think they can bolt it onto four rails and get a formula one? They'll be sure to get something quicker than a basic grocery getter alright, but it won't handle the curves under power so well without a little expertise and finess, which takes just a little more willingness to learn and get your hands dirty. In that sense I find your question somewhat loaded, albeit unintentionally.

Further, are you making such a statement with the ghostly image of a formerly modified module in your rearview ? If you so much as DC coupled your previous modules and make such a statement while the HG's are still AC coupled, you're comparing apples with crackers again.

The op amp on your old AD version though gave it a tightened, prominent bass, with hardened highs. The 5532 variant was grainier but far more natural and transparent, and the LM is closer to it, only a world improved. Signature is somewhat more relaxed, unconstrained, but can reach lower with greater authority, flowing musically, yet if the source material demands it, can be assaulting.

Assume the HG modules are extremely accurate. They will tend to sound as good or as bad as your implementation and as what they are fed with. This means layout, wiring, PSU, pre amp, and even recorded material.

While they will deliver rated power into dynamic loads if your speakers represent unreasonable loads they may go into limiting, which is something else to consider. In fact it's happened where "faultless" diy speakers intermittently drove the amps into protection under load, suddenly solved with the newer modules that offer limiting before a hard shut down.

My system is capable of extreme bass yet there are times when I find it's left wanting. Oddly it always comes after a few hours of playing bass light music, and the solution couldn't be easier, simply play a recording that actually has bass to begin with, and it starts to move the bowels just fine. It won't at all remind you of the compressed sound of a solide state pro audio amp though.

Maybe that's a flaw of high accuracy but in the long run it's a constant performer where for amps that tend towards artificially favoring the lows, compression or some other flaw, things always sound a little off when it comes to believability.

Aside from that there are certain things you can do to optimize LF power delivery and frequency response a little beyond ensuring your connections are in phase. You already mentioned one of them which is obvious enough. I suppose if you're willing to take the time in getting your hands dirty you could email me and I'd discuss them further, but I find little use in getting into it here.
 
#2,302
Classdphile, thanks for your detailed comments.

- Of course I did trace the path of the red-brown 0.68 uF caps. But double checking never hurts, hence the disclaimer.

- As you may be well aware, some solid state preamps don't have output caps anymore and may have a dc out in the range of a few mV. When you remove the input caps, this is multiplied by about 20 (voltage gain of the UcD). In addition, the offset voltage and offset current in the LM4562 of the 180HG is larger than in the AD8620; the maximum offset current is in fact significantly larger than for the AD8620 JFET inputs. This may or may not lead to a dc-voltage at the speaker terminals that exceeds what you are comfortable with. My limit here would be about 0.2 V DC. One of my moduls is actually at this limit (with the input caps still in place).

- The blue Bourns pot: Agreed, better don't touch unless you are dead sure which one is for the dc-offset. I probably shouldn't have mentioned that one.

- Bass response: I'm not talking here of small differences with a "ghostly image of a formerly modified module". I swapped solid state amps, UcD180 and Ucd180HG several times and the difference is significant. Yes, the 180 is dc-coupled and the 180HG is not. But the reasons for dc-coupling in the 180 and the 180HG are somewhat different: In the 180, you eliminate the electrolytics between the opamp and the modulator, which improves the sound as a whole. In the 180HG, the coupling caps have been shifted to the input and are non-electrolytic, so I would expect less of a sonic improvement by their removal. The 0.68 uF and the 100K input resistance yield a -3 dB point at 2.4 Hz. This seems a reasonable choice by Hypex and is certainly low enough; I fail to see how anything of significance in the bass response could be gained by dc-coupling. No apples and crackers comparison therefore.

- The 180 has somewhat less bass than my solid state amps, but I agree with you that the bass is tight and very well controlled; specifically at higher sound levels it really beats the solid state amps. At low listening levels, I prefer the solid state amps, since the 180 can sound a little thin. No mods have been made to the 180 except for the dc-coupling.

- The 180HG sounds much better in the midrange, where I perceive the detail, color and ambiance as first rate (better than with the 180). However, in my setup, deep bass is very much missing, specifically at lower listening levels (much less bass than with the 180). I have the version with HxR's, and no mods have been made so far.

This is just my impression and it applies to my setup and my system. I agree that this may seem kind of strange, but this is precisely the reason why I was asking how others perceive the 180HG bass response. I think the 180HG would be an excellent amp for me, if I could just somehow fix this bass-shyness. The speakers are BW802's and should not present a critical load.

If you have a 180HG, I would be interested at your impressions, also at low listening levels, and how you find it in comparison with the older 180.
 
#2,304
Javin5,

Do you know if your preamp/source has an output coupling cap and if so what size it is?

If it does remember the rule for caps in series. In other words, the total capacitance of both caps (preamp/source output and UCD input) in series will determine the roll-off corner frequency.

I can't offer any explanation, but I have observed a lack of bass with small coupling caps even when the maths suggests the roll-off should be insignificant. I try to use coupling caps that give me a corner frequency below 1Hz wherever possible (this approach also covers you for when you end up with two coupling caps in series).
 
#2,305
sharpi31:

The preamp I use is dc-coupled (no output caps); but your point about the caps in series is well taken. The only other cap coupling in my system is at the output of the cd-player.

In the meantime, I have bridged the input coupling caps of the 180HG. Bass response improved a little bit, but not nearly to what I have with SS-amps. I'm using the power supply modules and power transformers from Hypex.

It is difficult for me to determine if the sound quality in general is affected by the cap removal, since I can't do immediate switching for comparison. If there is a difference, it is not immedialetly evident. I would expect it to be small with foil type caps. The midrange quality, clarity and reproduction of details is very good anyway with the 180HG.

My intention was really to use transformer coupling (e.g. Lundahl) at the inputs, but I think that would probably yield a higher corner frequency than I could achieve with cap-coupling.
 
#2,308
bass response

Dear Javin

Those 802's are very nice loudspeakers but a rather "difficult load" for the amplifier. This has to do with the dynamic impedance which goes low (<3 Ohm) in two areas (bass and cross over frequency) when handling music signals.
IMHO the UcD 180 handles this nicely, so you have less "character" or "coloration" and a more "realistic" reproduction of your music.
 
#2,310
alternative to op amp?

Hi all,

I have a question about the input stage of these modules... Suppose I only use single ended connections to feed the UCD180, then can I eliminate the opamp input buffer and use a jfet buffer stage instead?
Are there any downside? is it be better to feed the pwm amp with a balanced signal anyway (even if the original signal is unbalanced)?

thanks!
 
#2,311
Re: alternative to op amp?

luigiinla said:
Hi all,

...then can I eliminate the opamp input buffer and use a jfet buffer stage instead? ...
Click to expand...

There has been post here with some tube input buffer. The point is just to provide enough gain (4.5 time gain). Look at hypex appnote on UCD gain. So if you find a discrete stage that has gain and can drive low impedance, it should do the job.
 
#2,313
thanks to both for your answers!

So how does the tube input sound compared to the opamp input?


Also, assuming that the UCD sounds better in balanced mode, why don't simply use a simple discrete buffer + a transformer in the input? This will also allow to eliminate the input capacitor..

Luigi
 
#2,314
luigiinla said:
thanks to both for your answers!

So how does the tube input sound compared to the opamp input?


Also, assuming that the UCD sounds better in balanced mode, why don't simply use a simple discrete buffer + a transformer in the input? This will also allow to eliminate the input capacitor..

Luigi
Click to expand...
Costs aside, would a transformer be that much better than a capacitor in this application?

Hans.
 
#2,315
It's very obvious that a transformer has its limitations. This is also clear when looking at phono (MC) inputs.
A capacitor also has limitations, but far fewer then a transformer. And in this case the excellent tube configuration (thanks to Jean Hiraga), combined with the capacitor, has certain advantages over a differential configuration with op amps.
BTW the proposed circuit has no input capacitors!
Having said this I realize that it should also be possible to configure a discrete circuit for this purpose and make it DC compensating for any off sets. But that would be rather complicated, think of some 30 - 40 transistors.

John
 
#2,316
Johnrtd, expensive solution that HD702( €2100). I know that design from the Netherlands.
I have build already 5 stereo amps with UCD180 and tubebuffer( 6n6p/ecc99 ) they cost no more then € 450,- for the parts and housing. And they play super.
I use only one capacitor in the tubestage( output cap).
 
#2,318
Johnrtd , for the symmetrical version i use also 2 tubes. 1 tube optimized for Z= 1.8 K and 1 tube optmized for z = 8k2. This because the UCD has behind the bufferopamp different impedance on + and - signalline. But that you also know.

I also use a delay( 20 sec ) to switch on the UCD so that the tubes can warm up and then switch on the UCD. This to protect the UCD against pumping.

About the price i use € 100,- max. for the tubebuffer and supply. The rest of the cost depends on the UCD's and supply you buy.

Ronny
 
You must log in or register to reply here.
Top Bottom