Any tricks to increase Quickie gain?

[Len]

I'm amazed at how happy I am with the two Quickies I have. One is being used as a low pass filter. The other is a headphone amp.

Gain is great for all decent sources. The only challenge I've run into is when feeding the headphone Quickie from my Blackberry using Moodio for radio feeds. Output is pretty feeble. Just changing the interconnect made a large difference in gain. I have the Quickie and the Blackberry volume controls maxxed out (the Blackberry gets warm).

With these settings, the only music I can listen to from the Blackberry through the Quickie is Jazz and only late at night when everything is quiet. No symphonies during the day. The sound through the Quickie and Senns just blows away the earpods directly from the Blackberry, of course.

Are there any ideas out there on getting more gain from the Quickie headphone amps? More voltage, a cheap SUT scheme, another tube? I'd like to stay with the 500 ohm tap for the Senns. I don't mind doing major surgery.

Thanks in advance for any suggestions.

 

[Len]

Here's where I'm up to in my thinking: A front end preamp for the Blackberry, built into the Quickie.

Perhaps a little Jfet step up (man I'll never get be able to get on an airplane with this thing now!)

Or maybe use my extra Specos (either that I got for this project or from my original Paramour 1's), with the 8 ohm as the input and one of the primary taps as the output? It would seem that I would be using them at a voltage too far below their design parameters.

How much can I step up the input before the Quickie clips?

 

[Wanderer]

Did you ever try the plate chokes?

The Hammon chokes upped the gain for me in a headphone Quickie.

The PJCCS is, according to Paul Joppa a post under that topic, supposed to give the same gain as the chokes. 

 

[Len]

Did you ever try the plate chokes?

The Hammon chokes upped the gain for me in a headphone Quickie.

The PJCCS is, according to Paul Joppa a post under that topic, supposed to give the same gain as the chokes.   

Thanks for the reply!

Yes, I have a pair of Magnequest chokes in there now.

Here is a link to what's in it:

http://www.bottlehead.com/smf/index.php/topic,218.msg1235.html#msg1235

The output trannies cut the gain by about 4 I think. It's not an issue with the Quickie, but with the Blackberry. Since the Blackberry doesn't put out anywhere near audiophile quality, I'm looking for a cheap way to boost its output.

I'm thinking of trying a Speco this weekend, but I'm kind of put off by trying to load it as a step up transformer correctly. I sure don't expect to put a scope on it and all. Just a quick and dirty patch is what I'm looking for. I'll see if I can find the specs on the Blackberry's output (impedance) before I do this.

Thanks again.

 

[Paul Joppa]

It would be easier if you could find some specs for output voltage and load (headphone) impedance, but I'd start with the Speco wired backwards. You might even use the 500 ohm tap for the Blackberry, 8K for input to the Quickie - that way you have more turns and more inductance loading the phone. If it works then we can talk about loading the transformer for optimum frequency response.

 

[Len]

You might even use the 500 ohm tap for the Blackberry, 8K for input to the Quickie - that way you have more turns and more inductance loading the phone. If it works then we can talk about loading the transformer for optimum frequency response.

Thanks. I cannot find any specs with the enclosed literature.

So you mean I should try it wired as an autoformer for stepup?

EDIT: I found replacement headphones for the Blackberry Tour to be spec'd at 16 ohms.

 

[Len]

It would be easier if you could find some specs for output voltage and load (headphone) impedance, but I'd start with the Speco wired backwards. You might even use the 500 ohm tap for the Blackberry, 8K for input to the Quickie - that way you have more turns and more inductance loading the phone. If it works then we can talk about loading the transformer for optimum frequency response.

Thanks for the suggestion. I wired it in as you said, and now have tons of volume on the Blackberry input! Nice and loud at one o'clock. I did the wiring on a new tagstrip, so I can change to any other tap if needed.

Sound is loud but seems sparse. Bass seems to be lacking, if not other frequencies as well. But it's working so far!

This is not a quick mod. I'm surprised how long it took so far. I mounted the 2nd set of Specos on standoffs over the 9V batteries, moved the 2.5mm jack there, put back the original input selector switch so that other sources would not flow through the autoformer, and put in the tagstrip so that there would be something to which the loading resistors (and caps if necessary) could be mounted.

This is one heavy Quickie! One pair of Magnequest plate chokes and two pair of Specos.

BTW, I used the same Specos that we use for output transformers for the SUTs, just so that if anyone else wants to mount cheap stepups and load them, we have a standard.

Blackberry earpod replacements are spec'd at 16, 17, and 18 ohms. There is a set of Bose that are spec'd at 32, but it seems that some Blackberries (sp?) don't recognize them when they are plugged in.

Have a great Thanksgiving dinner!

 

[Paul Joppa]

OK, now we're getting somewhere! I think this is the best solution, you just need better transformers. If you have plenty of gain now, try driving a higher impedance tap with the phone - less gain but more inductance, see what compromise works best.

For input transformers, there are many NOS classic transformers that are basically 500:500 ohms with at least  one side that can be wired for lower impedance, often 37.5 or 50 ohms. Most of them are designed for fairly small signal levels. I have some UTCs for example. Being too low an impedance for normal audiophile use they are often available cheap, and usually have decent magnetic shielding.

For output transformers, right now I'm thinking of Magnequest B7-8K units with 125 and 500 ohm taps; good for 'phones of 60 ohms or greater. I am talking to Mike at MQ about an official upgrade set; I just want to measure the BCP-16 grid choke as modified for plate choke duty before making a solid recommendation. Meanwhile most pro audio transformer makers will have a 10K:600CT or 8K:500CT unit that might work out. IMHO, I'd look for nickel cores to maintain inductance and linearity at small signal levels - but that's based on theory, we don't have a lot of experimental evidence yet to guide our choices.

 

[Len]

OK, now we're getting somewhere! I think this is the best solution, you just need better transformers. If you have plenty of gain now, try driving a higher impedance tap with the phone - less gain but more inductance, see what compromise works best.

For input transformers, there are many NOS classic transformers that are basically 500:500 ohms with at least  one side that can be wired for lower impedance, often 37.5 or 50 ohms. Most of them are designed for fairly small signal levels. I have some UTCs for example. Being too low an impedance for normal audiophile use they are often available cheap, and usually have decent magnetic shielding.
...

Thanks again. It sounds great now.

I moved the input to the 1K tap. Now I still have lots of volume, but the bass is fine! Quickie volume is at about 1 o'clock with the Blackberry at about half volume. So it seems to have gotten louder somehow. Maybe it's the increase in bass that makes it seem more comfortable at a lower setting.

I just caught most of the clarinet version of Rhapsody in Blue on Moodio.fm and the bass and top end were fine. Midrange was just a tad grainy, but that could be due to the format and/or the Blackberry. Today will be a bit of a busy day for the other Quickie, and the Blackberry with the Quickie headphone amp is sounding great, so I probably won't get to try the 2K tap today.

Thanks a million for your help. I would not have considered using the Specos as autoformer stepups. I still don't understand how that can work, sending a signal up a wire and having increase in voltage.

I'm done for now. Time to tweak the Quickie low pass.

 

[Dr. Toobz]

Did you ever get to try the 2k tap and compare it to the 1k? I'm thinking of buying another set of Specos for the same purpose, i.e., driving the Quickie with a portable player.

 

[Len]

Did you ever get to try the 2k tap and compare it to the 1k? I'm thinking of buying another set of Specos for the same purpose, i.e., driving the Quickie with a portable player.

Sorry, no. I haven't gotten to it. I thought I had come across a better SUT for it, but it was a telephone transformer and didn't have the bandwidth.

Keep in mind that the Specos are down about 3dB at 80 Hz, if I remember Doc's post correctly. It is probably worse when they are used at such low signal level as SUTs. It's not a problem if you are also using Specos as output transformers, but if you are not using an OT or for when you go to a better output transformer, it will be the weak link.

For now, I don't mind. The phones sound great to me, much better than the highly acclaimed Blackberry earbuds. The Blackberry's internet radio is probably the choke point in this setup, not the SUT. I haven't downloaded any hi res audio to store and play through the Blackberry yet, though. They should be of higher quality.

Even with all the qualifications, I am very pleased with the setup. You can't get more bang for the buck than using the Speco.

 

[machinehead]

Hi Paul, did you ever get a recomendation on a transformer for the Quickie?  Is this for headphones only or for increased gain all around?  I am looking
for increased gain using the Quickie as a preamp.  I have to use it at full volume to get a decent sound level, driving a pair of 200WPC mono blocks.

Thanks.

OK, now we're getting somewhere! I think this is the best solution, you just need better transformers. If you have plenty of gain now, try driving a higher impedance tap with the phone - less gain but more inductance, see what compromise works best.

For input transformers, there are many NOS classic transformers that are basically 500:500 ohms with at least  one side that can be wired for lower impedance, often 37.5 or 50 ohms. Most of them are designed for fairly small signal levels. I have some UTCs for example. Being too low an impedance for normal audiophile use they are often available cheap, and usually have decent magnetic shielding.

For output transformers, right now I'm thinking of Magnequest B7-8K units with 125 and 500 ohm taps; good for 'phones of 60 ohms or greater. I am talking to Mike at MQ about an official upgrade set; I just want to measure the BCP-16 grid choke as modified for plate choke duty before making a solid recommendation. Meanwhile most pro audio transformer makers will have a 10K:600CT or 8K:500CT unit that might work out. IMHO, I'd look for nickel cores to maintain inductance and linearity at small signal levels - but that's based on theory, we don't have a lot of experimental evidence yet to guide our choices.

 

[Paul Joppa]

A choke or current source load will increase the Quickie gain, nominally by about 6dB. It will, however, also double the output impedance, limiting the length of cable that can be driven comfortably.

An output transformer will decrease the voltage gain while it increases the available current - good for driving headphones or long cables.

 

[machinehead]

I typically only run 1 meter interconnects is that within the limit of a choke?

If so, will the choke filter frequencies?

 

[Paul Joppa]

No problem!

 

[machinehead]

Thanks Paul, I take it, that it won't filter frequencies?

Also I bought the PJCCS I take it that I should not use it in conjuction with a choke?

Thanks.

 

[Dr. Toobz]

If the output impedance is doubled using a current source or choke (I've used both, and am using the PJCSS right now), does that mean that the coupling cap values could/should be changed to a more ideal value? Or does this not matter?

 

[Paul Joppa]

For raw frequency response, the cap size depends on the input impedance of the following device. Usually 2uF is enough for a load of 10K or more (-3dB at 7Hz).

Larger is better because it keeps the impedance of the interconnect line lower so it is less likely to pick up interfering hum and noise from the environment. The point of diminishing returns is when the reactance of the capacitor becomes less than the source impedance. That in turn depends on what frequency you specify. For example, if 60Hz hum is the issue then you would use that frequency. This is my usual practice for line-powered devices like Seduction and Foreplay. For Eros, being a higher-end product, I doubled the output capacitance to get a tradeoff point of 30Hz.

Of course, larger means pricier, heavier, and often lower sonic quality, so it's a tradeoff. The optimum actually depends on the electromagnetic environment in your particular system, including that inside the system (the power amp for instance). If you want to use a different value, experimentation is your friend, all those internet gurus are not. (That includes me!)

 

[machinehead]

Thanks Paul.

Something I have question about.  You doubled the capacitance, you said which usually can reduce sonic qualities, but you also stated was a higher end product.
Did this just work better in that particular product?

 

[Paul Joppa]

No, we used a better quality capacitor. Sorry that wasn't clear. For the most part, capacitors of a given design (dielectric and conductor) will be similar in sonic qualities. But in general, the best materials often result in physically larger and/or very expensive capacitors, which at some point forces you to use a smaller/cheaper capacitor of lesser quality for purely practical reasons. This leads us to over-simplify and think "smaller value capacitors are better" - sloppy thinking, and I apologize for indulging in it.

The above is the reasoning that leads some to bypass a large value capacitor (often an electrolytic) with a small value, high quality capacitor (usually a film cap). In my experience, this often does not work because the interactions between the caps do more harm than the bypass does good, but sometimes it does help.