Bijou All Tube Futterman Headphone Amplifier - Page 144

that is friggin' cheap. i'd say that would call for a road-trip and a trailer.

Yes, I can create a couple extra Bijou badges for backlit purposes. Dont know if they will be glass front or acrylic front.

Fishski, sweet looking router!

Kanobs are ready to install through the front panel tonight. Thanks for lookin.

Going for the old Sansui/Marantz look.

keep me posted re: backlit Bijou badge. what are the dimensions? what final design/drawing did you settle on? i'm game as long as there's no mention of "balanced" on the badge . maybe i'll need 2 badges in case my SE chassis clones another down the road. Bijou mono-blocks for balanced ?

Alex:

I'm wondering why some designers put a very small capacitor (i.e. 10pF) in parallel with the negative feedback resistors? I've seen this on the Leech amplifier for example. I'm curious what the function of this capacitor is, and in what types of applications it is recommended. Can it be used to reduce the output capacitance needed, for example, or perhaps used in lieu of a DC servo?

Inverting Amplifier Feedback.

that may shed some light.

yeah, nice link answers my questions, except for why doesn't everyone do this.

Fishski, what is Discount Steel's eBay store (originally found them that way, but can't atm)?

Oh, yes, still loving the Bijou - warmed up, and ready to go head to head with a balanced beta22 in two weeks. We will see what is what, with neutral observers weighing in.

Now that sounds like fun.

What did you finally settle on for an attenuator?

This should probably be done more often than it is, but many casual designers may not understand the phasing problem around the feedback loop.

There are usually two ways that the HF instability is removed. The closed loop compensation is one way. There is also open loop compensation which is usually applied at the main voltage gain stage (in many transistor amps, the VAS). This cap causes the gain to drop quickly above the chosen corner frequency so that the amp just doesn't have any gain with which to oscillate. Then, if needed, the close loop compensation is also added.

In a sand amp using the most typical topology you'll see the open loop compensation cap called Cdom and it sits between the base and collector of the primary gain device (or gate/drain). At HF this cap shorts the base to collector steadily reducing the gain of the transistor (or mosfet) with increasing freq.

Thus, for the HF loop to be a problem the amp actually has to have bandwidth high enough to amplify in that region. If the HF bandwidth is not high enough the closed loop compensation may not be needed. The Bijou's HF response is not super high because of the tubes, so the NFB issue is much less of an issue. However, I actually did test this condition and was able to make the Bijou oscillate by lowering the value of the first grid stopper. The 10k value puts the amp well away from an unstable regime without additional compensation.

Does this help?

i had no idea they had an eBay store. i've only jibbed around on their www.

yes, this definitely helps. So, with regard to HF LPF rolloff grid stopper resistor solution compared to compensation capacitor from the plate to grid in a tube output stage, what are the pros and cons of addressing oscillation these two ways?

It seems that there are 3 different knobs here to address this problem, and it's hard to know whether one turns up a single knob high enough to solve the problem, or turn more than one knob up slightly...

Also, how did you test this condition on the Bijou, via SPICE or the actual circuit? It seems that this is one of those cases where SPICE might not tell the truth, or at least that happened with a solid state amplifier I built..

I doubt you could simlulate the bijou in spice.

My guess is alex put a low gridstopper in, and scoped it.

TKD P2511 (the PartsConnexion 'Alps' taper), which sounded like 'sand' initially, but really straightened out after about 100 hours or so, most of which was after about 20 hours... but, initially, they sounded like poo.

I am scared beyond reason that Marc poked in here... after being at my house, and not getting a listen to the Bijou, at all...

'Course, we had electrostats to fry...

I can't wait to read those impressions.

I will give a brief spoiler of balanced Beta22 vs balanced Bijou. It will be brief because I highly respect both designers and applaud them for the results of their hard work. And having enjoyed the results of their work so thoroughly, it makes it difficult to find any negative aspects for the review. So, when I attended the Burning Amp event, the balanced Beta22 was hooked up to some quality speakers for about an hour. It filled the air with some of the richest and effortless sound, that I have ever heard. Incredible detail and soundstage. The music was extremely lively and timing was right on the money. There was a nice balance between treble and bass. With plenty of tight bass on hand. Strings/violins resonated realisticly. The balanced Beta22 should not be used with speakers without large heatsinks in the output stage however, because of the large power dissipation there. It is designed as a headphone amp. In comparison, the balanced Bijou with Pabbi's HD650 headphones, had an extraordinarly spatial sound with deep bass, prominent but smooth mids, and great extension with the high frequency range, giving vocals and acoustic guitar a spine tingly timbre. There is a great presence of being there with the artist. It is engaging and you get lost in the music. It is not fatiguing, and you want to listen for hours(usually 4 for me). Both amplifiers run with a fair amount of heat, and it is necessary to keep that in mind when building either. I will leave it to Pabbi to do a head to head on phones. ;-) Edit: I would love to hear the balanced Bijou with some super efficient speakers, but probably never will due to the price and the fact that the load impedance would be too low at only 8ohms per channel, effecting linearity and current consumption.