ImPasse Preamplifier

Thanks all. A visit to a local trophy shop for the lettering. Very reasonable, plus I learned of another technique they use for lettering most any object, UV printing. Capable of printing in different colors and in the case of lettering on this type of project, should last for a lifetime. I'm going to try it out on the next project.
Definitely will be giving these pres a listen through the F4s and Mofo for comparison.
David
 
OT Quad Resistors

We had discussed Quads earlier in this thread. For those interested, I have put up a thread that pulls together performance data for Quads compared to other resistor technologies. At low resistance values, it turns out that power distortion is the biggest area of improvement. That's less so at higher nominal resistance, but still an interesting technology.

The thread is in the Parts Forum under the title;

Quad or Series/Parallel Resistors as an alternative to traditional power resistors

Jac
 
This is one person's thoughts on the AKSA-Lender, the Nelson BA-3 and the Yaniger ImPasse preamps. The amps being used are a pair of bridged Pass designed F4. The F4 amps need a preamp with a larger output voltage than most preamps can deliver. The three listed above are capable of the necessary output.

My first DIY build began in 2014 was the BA-3 followed by one F4. I liked that combination so well I build another F4. The ImPasse was something I wanted to try and finally late last year purchased the boards and input transformers from Jack. Then Hugh delivers his AKSA-Lender preamp to the community and I wanted to try it also with the F4 amps.
Please keep in mind these are my thoughts using this three preamps and your listening experiences will vary, due to variances in the builds, your listening space and choice of speakers.

My BA-3 build uses a single toroidal transformer into a Salas shunt regulator. The BA-3 has genuine Toshiba transistors and I am using 10uf Mundorf SIO caps on the outputs with Goldpoint pot and selector. Using the P3 adjustment, the second harmonic has a negative phase. The AKSA-Lender was built from the BOM listing from that thread. I chose R9 for a more dominate second harmonic. Finally the ImPasse was also built using the BOM from that thread. I did do some tube rolling, started with Electro Harmonix and really didn't like the sound. Finally ended with some NOS RCA and Amperex tubes.

I am using a Pink Floyd song from their 1973 album "Dark Side of the Moon". That song is " The Great Gig in the Sky". The instruments are a piano, Hammond organ, pedal steel guitar, bass, and drums. The main vocals are supplied in an almost orgasmic way by Clare Torry. Her wailing reveals much about the preamps. Also I listened to the live version of Dave Brubeck's "Take Five" from the album "Concord on a Summer Night".

BA-3 preamp
A nice wide sound stage, with the pedal steel guitar coming more from the right side. Good depth with the vocal and drums coming from the rear. Cymbals are crisp. In the second set of vocals distinct sounding. The piano sounds very smooth and rounded. There is a nice sonic balance with the bass.

AKSA-Lender preamp
Also a wide sound stage. The pedal steel is not as defined as the BA-3. Cymbals have a nice sizzle. The Hammond organ is distinct and the vocals are more forward sounding. Drums are also very defined.

ImPasse preamp
Another wide sound stage. The piano and pedal steel are well defined. Vocal appears from upper left with the bass more defined. Cymbal also with nice sizzle sound. In the second set of vocals there is a nice balance with the piano. For me the vocals were not as clearly defined as the two solid stage preamps.

Finally the question would be which is best/sounds better, etc. For me there is a personal enjoyment when listening with each preamp at different times with different music. Ok, my order of preference would be the BA-3, AKSA-Lender, and ImPasse. I believe both solid state preamps have a slight edge over the tube preamp. Finally if you want to look at cost and versatility then the AKSA-Lender has both of the others beat for price and convenience. In my opinion it offers the most bang for the money.

So final thoughts, any of these preamps would be great with the Pass F4 amps. My wife and I for the past twenty years have been involved with retired racing greyhound adoption. In that adoption world, there is a catch phrase that is applicable to the DIY audio world. It is "they are like potato chips, it's hard to stop with only one". True for the greyhounds and so true for the DIY audio world. All I can say is build and enjoy.
Thanks
David
 
Yes, the resistor you linked is the one I used.



One advantage of using SMD or MELF resistor for this kind of quad construction is that you can keep the resistors close to the pcb and minimize inductance increase. I would guess that my homemade quads don't have much more than a Panasonic ERG (see post 1097).



One other interesting thing here is heat rejection. As an experiment, I soldered a 15k, 3W Dale CPF (tempco 100 PPM/C) to a proto board and ran it at about 1 W. I gave it 15 minutes to stabilize and measured temperature with an IR meter. I then repeated the experiment with the homemade 15k quad resistor. The results were that the Dale CPF temperature was 40 to 45 deg C above ambient and the quad was about 10 deg C above ambient. Of course, the CPF has much less surface area to transfer heat to the air. I was kind of surprised that a 3W resistor would be so close to it's 70 deg C derating temperature with only 1W.



Heat rejection is an important part of this because of power modulation distortion. The idea is that a hot resistor has a different resistance than nominal and that change in resistance changes the output of the circuit. That change of output is distortion. Since music has quiet and loud passages, the resistor will change resistance with the music and add non-linearity to the distortion.



Power modulation distortion is a function of the tempco and the temperature change as a function of power. Going back to the 3W Dale CPF and the Susumu quad, the CPF has higher tempco and higher temperature so it should have worse power modulation distortion. Relating back to the Panasonic ERG, it has much more surface area than the Dale CPF, so it would likely have a temp change with power in between the Susumu and the Dale but it also has a tempco of 300 PPM/C for the 3W and 350 PPM/C for the 1W. I don't know where a 1W Dale CPF would come out in a quad, but one of the advantages of SMD/MELF are that they are designed to transfer heat to the pcb, so its possible they may have a lower temperature even in a quad.



One last factor for the quad. Distortion in resistors is said to be mainly a function of voltage. In the quad, each resistor is operating a 1/4 the voltage of the total array. The theory says (per Groner) distortion of a quad should be 12 dB lower than a single identical resistor. I read an EE forum that suggests you can see this affect in Spice modeling, although I haven't tried it. That said, I'm not sure it will show up in measurements. X was working on the AKSA Lender pre and having trouble with a bit of 3rd harmonic distortion. On my suggestion, he tried a quad (without board) using thick film SMD resistors. The measurements did not improve.



Why not a Dale RS Wirewound?

Something like 20ppm for that series.

In an area where inductance would be an issue I believe Morgan Jones found wirewound inductance negligible beyond 18K or thereabouts.

Of course you could always use a non-inductive wirewound like the Dale NS or Mills.

I don’t think the paper you are referring to had wirewounds in the mix.

What’s the value in question?

PS your experience with the CPF series is quite opposite to mine, especially the larger 3W.
 
Last edited:
Why not a Dale RS Wirewound?

Something like 20ppm for that series.

In an area where inductance would be an issue I believe Morgan Jones found wirewound inductance negligible beyond 18K or thereabouts.

Of course you could always use a non-inductive wirewound like the Dale NS or Mills.

I don’t think the paper you are referring to had wirewounds in the mix.

What’s the value in question?

PS your experience with the CPF series is quite opposite to mine, especially the larger 3W.

If you are interested in a spec based discussion of this issue, the thread at the link below goes into more detail with some very expert people taking part. I don't claim the quad resistor is the only power resistor choice, but it is an interesting and new approach that I have found to be successful.

Quad or Series/Parallel Resistors as an alternative to traditional power resistors

The idea for the quad resistor is not mine. It has seen various uses and mentions, but I found it in the paper linked below, under passive components.

http://www.nanovolt.ch/resources/low_distortion_oscillators/pdf/low_distortion_oscillator_design.pdf

Wirewounds are clearly the closest option meet specs discussed for the quad resistor. Even at 20 ppm, cylinder form wirewounds don't have the heat rejection of the quad design, so power modulation distortion won't be quite as good.

Subjectively, wirewounds are a matter of personal choice.

You are correct, the paper I referenced didn't test wirewound resistors because their laboratory application couldn't stand the inductance. As always, sensitivity to things like inductance are dependent on the circuit.

The value in the Impasse is 15k. I have also used the quad in a different amplifier in a 0R33 value with success.
 
Resurrecting an old thread, but I recommended the Impasse to someone last week, and I thought I'd take my own advice.

I ran into a problem almost immediately, however, when I started a Mouser parts order. (Let me start by saying that I hate to order parts, in part because my crappy laptop doesn't respond well when trying to navigate the search results, either Mouser's or D-K's.)

At any rate, to my tired eyes, there is no such thing as a 12k5 2W resistor. Actually, as I think back over the last couple of hours, I believe that at one point there appeared a Dale 2W or 3W, at about $3/pop, but not in stock, and with a lengthy backorder time. But could it be true that there's only one? I tried specifying a 12-13 range too, but that produced nothing usable, unless I wanted to order 5,000 of them.

I could parallel and achieve this, but there's usually something untidy about the result when I try that. Any suggestions? TIA - Pat
 
At any rate, to my tired eyes, there is no such thing as a 12k5 2W resistor. Actually, as I think back over the last couple of hours, I believe that at one point there appeared a Dale 2W or 3W, at about $3/pop, but not in stock, and with a lengthy backorder time. But could it be true that there's only one? I tried specifying a 12-13 range too, but that produced nothing usable, unless I wanted to order 5,000 of them.

t

How many do you need? I have about 100 of the Vishay 12.5K 3W 5% Silicone -- not ROHS compliant. The new variety are $1.29 each (Both Avnet and DK are out of stock.) RS02B12K50FS70
 
I ran into a problem almost immediately, however, when I started a Mouser parts order. (Let me start by saying that I hate to order parts, in part because my crappy laptop doesn't respond well when trying to navigate the search results, either Mouser's or D-K's.)

At any rate, to my tired eyes, there is no such thing as a 12k5 2W resistor.
I could parallel and achieve this, but there's usually something untidy about the result when I try that. Any suggestions? TIA - Pat

Were you specifically looking for thin film or something? I searched Mouser US and found 11 options of metal oxide, all in stock and available in quantities of 1. The Panasonics that the Pass guys like were $0.24 to 0.50 and the KOA Speer were $0.14.

As an aside, I have used both the Panasonic and the KOA Speer with some success, although I admit that I prefer other resistor technologies.
 
Were you specifically looking for thin film or something? I searched Mouser US and found 11 options of metal oxide, all in stock and available in quantities of 1. The Panasonics that the Pass guys like were $0.24 to 0.50 and the KOA Speer were $0.14.

As an aside, I have used both the Panasonic and the KOA Speer with some success, although I admit that I prefer other resistor technologies.

Interesting. I must not know how to search. I just checked, and the drop-down menu for MO resistors does not list 12k5, just goes from 12k to 13k. Any brand, any power. Using the sitewide search window, I find 55 12k5 (actually 12.5k, searching for 12k5 returns fewer possibilities) resistors of any type, with only those Dales in that resistance and adequate power rating. I don't doubt that there are more, but I can't find them.

Jack, I'll pm you. Maybe there's more stuff that you can pull out of the warehouse for me. ;)

Thanks guys.
 
@tubesguy,

No, the fault is mine.

I checked and I had clicked on 13k. You are correct, there is no 12.5k choice. I'm just click happy I guess. Sorry.

I found 12.7k, 2W and 3W in a Vishay CPF series, but its $3 and I am not a fan. In my opinion, these run hotter than they should due to their small surface area and may exceed the temperature of the power rating.

So the real question is, can you run 12k or 13k where there are reasonably priced MO resistors?

The only other thing I can suggest is truly out of the box. You could do a quad resistor or series/parallel using 4 SMT 1W resistors for a combined power rating of 4W. Something like these might work.

CRCW251212K7FKEG Vishay / Dale | Mouser

I would be happy to provide you with KiCad or Gerbers for a little pcb to mount them. At Osh Park, you can get 3 of these little guys for under $4. The pcb acts as a heat sink, so they run cool. Or you can make your own out of copper clad board. OK, enough weird thinking.
 

Attachments

  • KRL FE R3.JPG
    KRL FE R3.JPG
    220.7 KB · Views: 462