ImPasse Preamplifier

wirewiggler said:
I have my power supply finished point to point, I went with the Temp compensated version. What would be a good load resister to test the power supply with ?

Bill

Read the article where it specifies the current load and voltage output of the PSU, then use Ohm's Law to calculate the resistor value, then again to calculate the power rating.
 
hmm, 8750 ohms by 14 watts at 350V nothing laying around the bench like that . I could hook it up to sparky the dog and see how high she jumps but I have grown kind of fond of her. I want to make sure the circuit is operating and regulating properly, I will bring it up on a variac how would you test it ?

Bill
 
Thanks Sy
 

Attachments

  • p1000195.jpg
    p1000195.jpg
    97.5 KB · Views: 916
Well, the regulator boards will be sent out soon, so stay tuned.

I didn't get to work on the ImPasse as much as I'd like to over my vacation, but I did make a little progress. Problem at this point is I'm at 249 of 250 available pins (without paying for an upgraded license), so I'll have to proceed carefully with regards to adding features and what not.

So far things are singled sided, but I have a few problem child connections to deal with so it may not stay that way. Board size is 7.3" x 6.2", with plenty of room for caps :)

This is just a first draft, but wanted to open things up for comments and suggestions! For example, I realize things are pretty jam packed around the tubes, and that can definitely change if people are worried about heat... the floating connector is for a CCS bypass that can also be used with an ammeter to dial in the CCS using the pot located on the board to experiment with various operating points, as well ensure good device matching. Without further ado:

An externally hosted image should be here but it was not working when we last tested it.


Better quality version is located at:

http://lh3.ggpht.com/_8RHE4QJC4jI/SnifbWQlf4I/AAAAAAAAAYs/CXzEfBBL7CY/Impasse_1.0.jpg
 
oh one last question :)

I plan to have separate B+ connections for each channel. At the moment, there are two seperate star grounds, one for each channel. My question is, should these connect at the point about the ground loop breaker, or perhaps keep things separate and have a ground loop breaker for each channel that connects to a common earth, or does it really matter? :)
 
I connected all stars to one central point, PS gnd to that point, then the groundbreaker between there and the chassis. You could also have a nice, thick bus starting from input grounds to output, to PS.

I'll confess to only having one B+ feed. The PSR of the gain stage is ridiculously good, and the splitter-output stage is handling 20V signals.
 
The power supply is only putting out 170V with 350 in. I put it on my test power supply (max out 250V) and here are a few of my readings input of U1 is 133V collector of U2 and U3 250V emitter 135V base 135V any ideas ?

Bill

Check your pass transistors, polarity of D1 as Jan mentioned.

The output of the regulator should be ~ (1 + 8.2) times the ratio of R6/R5 (R7,R6 in LB47). The 8.2 is the value of the LM329 plus the LM317's own reference voltage.

If you have a good idea of how much work the regulator is going to do you can change the value of the resistor before the LM317. I have shorted this thing and blown out the Darlingtons.
 
a brief update. I ran out of pins in my PCB software, so I had to put the project on hold for a bit. I finally had enough in my slush fund to pay for the upgrade.

I've decided to try and PCB mount the input and output jacks, as well as tailor the PCB to fit in an off-the-shelf extruded case. The PSU will be in a separate enclosure, more like a "floor wart". The case I have chosen for the preamp section is available at Lansing Enclosures. The basic enclosure is a Micropak C, specifically part CW2B90-B1B2-B1B2. It retails for around $40. I also will be accommodating a custom enclosure from Front Panel Express, which will be a bit pricier, but is more along the "paint-by-numbers" assembly style.

Anyways, just didn't want anything to think I forgot about this project, so look for an update in two weeks or so :)
 
Make sure that you leave enough space for R4 and its sister -- I measured the temperature on these at over 50 deg C. R11 will also cook a bit, but 1W is sufficient as SY illustrated.

It's not necessary to put C106,7 on the PCB -- best these ceramic discs are soldered right to the pins of the octal socket and attach to the ground "locus" with as short a lead as possible.

At the moment I am using my LR8N3 regulator with an MJE3439 pass transistor -- waiting for the power supply boards to come back so I can see how they work.

I cut the trace from ground to the shield of V2 -- I was uncertain how it would take to the fact that the filaments are lifted off ground -- ymmv
 
Last edited:
The ceramics' working voltage doesn't need to be terribly high, but most RF ceramic discs I have are 600V. I would think that a 200 or 250V rating would suffice. The main function is to get the diode hash and other noise on the heater line down to ground, where it can't bother anyone.

I've started a rebuild of mine (it was damaged in two successive cross-country moves) with Jack's PCBs.