Need a pair of eyes on a spectra sonics 101 vero layout

[digitana]

yes I also saw the newer 101 cards. Finding 5pf polystyrene is not so easy  but I  have some silver mica's I can try.
I also have some NOS Fairchild 2n3566 and 2n3645  but I havent checked their beta's so dont know if I can make decent matched pairs or not...

Now if only someone could  reverse engineer the  Spectra Sonics  502  passive eq  that was designed to work in the feedback path of the 101/110 and make a drawing ...  :0)

 

[orangechili]

Hopefully you can get some decent pairs out of them. Sometime this week i'll knock together a bipolar 24v supply and test out the 110 layout.

I can't seem to find any schematics on the 502 eq modules… It'd be interesting to see some more info on that one for sure!

 

[orangechili]

Alright, I knocked together a bipolar 24v supply, built up the spectra sonics 110 vero and it works well It is verified as well.

Here's my observations, keep in mind I've only built 1 board of each so I'm basing this off a small sample size….

-The 110 is quitter in operation, dead quiet to be exact. Love it! I used Edcor wsm 600:600 transformers for the ins and outs on both pres cause I have a bunch of these lying around. Eventually want to try the Triad a67-J wired as 150:600 for the inputs as people seem to like that best.

-The scratchiness of the pot caused by the dc isn't as noticeable on the 110, much less offensive

I notice the Q4 and Q5 transistors can kick off some heat at times on both the 101 and 110(hot if you touch them). I would recommend the bigger bodied original 2n3566 transistors because of this instead of smaller bodied modern replacements. I did match my pairs to well under 2mv as well and tested every resistor before going on the board. I could have used higher quality resistors and caps for sure but wanted to prototype these and not make a custom order for these parts just yet. I have been running these for an hour straight and no smoke. 

If any owners of racked up 101 or 110's can comment on how hot the last 2n3566 pair runs for them that would be helpful for sure! I'm wondering if this is normal for their operation to just run a little hot or if even more careful component selection and matching is the key.

 

[Whoops]

Alright, I knocked together a bipolar 24v supply, built up the spectra sonics 110 vero and it works well It is verified as well.

Here's my observations, keep in mind I've only built 1 board of each so I'm basing this off a small sample size….

-The 110 is quitter in operation, dead quiet to be exact. Love it! I used Edcor wsm 600:600 transformers for the ins and outs on both pres cause I have a bunch of these lying around. Eventually want to try the Triad a67-J wired as 150:600 for the inputs as people seem to like that best.

-The scratchiness of the pot caused by the dc isn't as noticeable on the 110, much less offensive

I notice the Q4 and Q5 transistors can kick off some heat at times on both the 101 and 110(hot if you touch them). I would recommend the bigger bodied original 2n3566 transistors because of this instead of smaller bodied modern replacements. I did match my pairs to well under 2mv as well and tested every resistor before going on the board. I could have used higher quality resistors and caps for sure but wanted to prototype these and not make a custom order for these parts just yet. I have been running these for an hour straight and no smoke. 

If any owners of racked up 101 or 110's can comment on how hot the last 2n3566 pair runs for them that would be helpful for sure! I'm wondering if this is normal for their operation to just run a little hot or if even more careful component selection and matching is the key.

Nice one.

Just to let you know that neither the 101 or the 110 were design to be used with the pot. The feedback resistor is supposed to be fixed, set and forget. Gain should be changed by the use of input Pad or output pad.

Here are the recommendations of Richard Guy.  Richard worked in Spectrasonics and has 50 years of experience with this gear.  He is probably the most knowledgeable person on Spectrasonics designs in this planet, much similar to what Geoff Tanner is for Neve stuff, here it is:


"Gain Adjustment

The Spectra Sonics pre-amplifiers are optimum at 40 dB of gain. It
is possible to adjust the gain by changing the feedback resistor according
to the following chart:

35 dB gain = 5.6 k resistor
40 dB gain = 10.7 k resistor
45 dB gain = 20 K resistor
55 dB gain = 42.2 k resistor

Resistors should be low noise types, preferable mil-spec metal film
types. RN60D is a good designation to look for.

Installation of a gain-adjust potentiometer is not recommended. It
will compromise stability of the amplifier as well as adding noise.

Switching feedback resistors is also not a good idea. At the very
least, it would require a system of constant value feedback resistor
bypassed by higher value parallel resistors to ensure that the feedback
loop never became open. Furthermore, changing the resistor
will be noisey, so switching while "on" would not be recommended.
I suggest you find a gain setting that works for your lowest output
microphone, then utilize input pads to make the adjustment. This is
superior in performance and stability, and does not add noise or distortion.
"

 

[orangechili]

Nice! I was planning on implanting the input pad like JLM audio's diagram in the other spectra sonics thread. I just used the feedback pot as a quick and dirty way to test the vero layouts. There is actually a place on both layouts to implement the feedback resistor for a fixed value.

The next step for me is to build a few more boards with the RN60D resistors and better capacitors and compare it functionally and sonically to my existing boards.

I've read somewhere else here  that these pre's kick off a bunch of heat. I'm wondering if Q4 and Q5 are the source of that heat as i've experienced? If anyone with racked spectra sonics pre's can comment on how hot those transistors run that would be particularly helpful.

I'm hoping that people can get some use out of these layouts, it really is a simple layout. The hard part is obeying the ground scheme, sourcing the rare transistors and matching them.

 

[Whoops]

Nice! I was planning on implanting the input pad like JLM audio's diagram in the other spectra sonics thread.

I think the input pad as described in the JLM schematic is not a good idea. It's used before the transformer, a dual gang Pot will have too much tolerance in each gang and will mess around with the common mode rejection. Pots can have tolerance from 10% to 30%.
But let me know if someone has a different view on this.

You will be better with this kind of arrangement (Richard Guy recommendation):

 

[Whoops]

Also the wiring for 101 and 110 amps from Richard Guy:

 

[digitana]

wow thanks for all this great info. Looks like I am going to have to build some 110's as the ones I thought I was going to get didnt come through :0( . I have  some  of the  original large bodied  fairchild transistors..so hopefully they wont run too hot.  Lets hope I get some luck matching pairs . I have a  triad  A67-J input transformer which I will try...

Thanks for the input pad idea. I was also thinking  to keep the gain  fixed because a pot in the feedback loop might  not be ideal. ..but wasnt sure what would be the best pad option...

 

[digitana]

PS: I joined the Spectra Sonics Archive site  which I think  Richard Guy runs ?  I think its still early days and not much in the way of schematics on there..but  lots of  other  literature.

 

[Whoops]

PS: I joined the Spectra Sonics Archive site  which I think  Richard Guy runs ?  I think its still early days and not much in the way of schematics on there..but  lots of  other  literature.

Yes I think he runs it.
He is a great person, really helpful and knowledgeable.


He also regularly sell 101 and 110 Boards and Kits on Ebay:

http://www.ebay.com/itm/Spectra-Sonics-2-Channel-101-Pre-Amp-Mini-Kit-with-Triad-Transformers-/261865276959?pt=LH_DefaultDomain_0&hash=item3cf863061f

 

[orangechili]

Those kits are great, especially as sourcing the original transformers/boards can be very time consuming! Building from scratch definitely turns into a labor of love as well with having to source obsolete transistors and diodes then matching transistors, etc. The circuit looks deceptively simple on paper...

 

[digitana]

yes they are very nice kits ..but the high cost of uk customs/vat make it prohibitively expensive  to purchase  ..eg for those dual 101 kits on ebay I would be charged $150 import dut and vat alone    :'(

 

[mjrippe]

I racked a pair from one of those kits (with the Bauer outputs) for a client.  Did everything just as Richard said and they turned out great.  My client raves about them!

 

[Whoops]

Will just leave here an original  110A board traces scan, in case anyone wants to etch at home similar boards.

 

[Whoops]

component side

 

[Whoops]

for the parts you can get transistors from  Small Bear Electronics and from Ebay also:

2N3566

http://smallbear-electronics.mybigcommerce.com/transistor-2n3566-fairchild/
http://smallbear-electronics.mybigcommerce.com/transistor-2n3566-work-alike/
http://www.ebay.com/itm/10PC-NEW-FAIRCHILD-2N3566-TO-105-PACKAGE-30V-200mA-NPN-BJT-TRANSISTOR-/131449356552?pt=LH_DefaultDomain_0&hash=item1e9afe0108


2N3645

http://smallbear-electronics.mybigcommerce.com/transistor-2n3645/

http://www.ebay.com/itm/2N3645-FSC-1965-67-GOLD-GLOB-TOP-TRANSISTOR-VINTAGE-/370458576000?pt=LH_DefaultDomain_0&hash=item56410d4480



I can't find sources for the 1N416A diodes, do you know any source?
Or is it a typo in the schematic and it suposed to be 1N461A?

 

[orangechili]

I used 1n461 diodes and they look exactly like the glass style ones i've seen on other older boards if that helps.

 

[digitana]

Thanks again whoops for the great support . I have a cnc  engraver which I have also used to mill pcb's rather than etch. For simple artwork like the  101/110 it  should be fairly easy to do... so I will give i t a go. The software I use can create Gcode (used by cnc machines)  from original artwork by tracing .

 

[digitana]

ps: I  bought a lot of 1n461  from Surplus Sales of Nebraska at $0.2 each. They are old types with colour coded bands (yellow, blue brown for 461). However I then realised the schematic has 1n461A which has a higher forward current(100ma) ..compared to the 15ma of the 1n461 so not sure how the  1n461 will perform :0(

 

[Whoops]

ps: I  bought a lot of 1n461  from Surplus Sales of Nebraska at $0.2 each. They are old types with colour coded bands (yellow, blue brown for 461). However I then realised the schematic has 1n461A which has a higher forward current(100ma) ..compared to the 15ma of the 1n461 so not sure how the  1n461 will perform :0(

I dont have any schematic showing 1N461A diodes do you have one?
My schematic shows 1N416A, I would like to know if that was a typo error or if thats really the Diode used.