M2 inspired balanced output preamp idea

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
An Aleph30 is sitting on a shelf awaiting a balanced output preamp.

Also sitting on a shelf are the parts for an M2 and some of the pcbs in the attached pictures. The buffer PCBs were acquired from eBay. They have a ridiculously wide frequency bandwidth.

The idea is to have one buffer channel driving an M2 Edcor transformer and a pair of buffer channels buffering the balanced output of the Edcor transformer.

Today I set up an Edcor PC600-15k using a sig gen and a scope. Driving 1/2 of the 600 ohm input and measuring across the 15k winding there is a voltage gain of approximately 9.5 and a -3dB bandwidth of 18Hz to 25 KHz with the output unloaded and the input driven by a 50 ohm sig gen.

Interestingly, the voltage gain is nearly identical to a BA3 preamp that I am finishing for a long-time friend. The parts cost for the M2 inspired preamp will be much less than the parts cost for the BA3 preamp.

Any comments/questions/suggestions are welcomed. Execution of this idea will have to await finishing the BA3 preamp and acquiring or re-purposing some enclosures.
 

Attachments

  • 04.jpg
    04.jpg
    136.4 KB · Views: 793
  • 05.jpg
    05.jpg
    51.1 KB · Views: 816
watching closely..
I have been blundering about using this buffer in Spice (and largely unsuccessfully transcribing from virtual to real). Each board is a two channels and my intent was to use one channel to drive the autoformer and the second as a buffer to drive some current for a big 2Sk180 configured as a choke loaded source follower. A kluged buffered version of the SIT-3 Nelson posted. Its on hold pending more and understanding.
 
care to cover that story with some schmtc ?

can't grasp are you talking about autoformer or xformer arrangement

though , through some fun I had making Iron Pre and Iron Pumpkin , I have a strong feeling that you need buffer(s) just in front of iron , no need after

Not autoformer. The autoformer gives you gain of N + 1. For me, N by itself is good enough... and ... I want the input isolated from the output.

So.. it is very simply

unbalanced input => buffer pcb => Edcor => buffer pcb balanced => output

I agree that output buffer is not required. Output buffer is only required if there is a need to drive a low impedance load.

You could call this an unbalanced to balanced converter with voltage gain.

One note is that there will be zero negative feedback which will make this preamp Zen-ish.

You could do this idea with B1 buffers.
 
watching closely..
I have been blundering about using this buffer in Spice (and largely unsuccessfully transcribing from virtual to real). Each board is a two channels and my intent was to use one channel to drive the autoformer and the second as a buffer to drive some current for a big 2Sk180 configured as a choke loaded source follower. A kluged buffered version of the SIT-3 Nelson posted. Its on hold pending more and understanding.

These buffer PCBs work well. You could apply + / - 40V with no problem making them useful with an OPS power supply.

The cost of finished working PCBs is so low that it is silly to acquire all the parts and assemble yourself. Unless that is what you love to do.
 
Member
Joined 2011
Paid Member
Calculate the output impedance of the preamp. How much capacitance can it drive without damaging the HF response? How many feet of interconnect cable is that? What happens when you connect a solid state power amp whose input impedance is 10K ohms, at the far end of a 5 meter interconnect cable?
 

PKI

Member
Joined 2011
Paid Member
An Aleph30 is sitting on a shelf awaiting a balanced output preamp.



Also sitting on a shelf are the parts for an M2 and some of the pcbs in the attached pictures. The buffer PCBs were acquired from eBay. They have a ridiculously wide frequency bandwidth.



The idea is to have one buffer channel driving an M2 Edcor transformer and a pair of buffer channels buffering the balanced output of the Edcor transformer.



Today I set up an Edcor PC600-15k using a sig gen and a scope. Driving 1/2 of the 600 ohm input and measuring across the 15k winding there is a voltage gain of approximately 9.5 and a -3dB bandwidth of 18Hz to 25 KHz with the output unloaded and the input driven by a 50 ohm sig gen.



Interestingly, the voltage gain is nearly identical to a BA3 preamp that I am finishing for a long-time friend. The parts cost for the M2 inspired preamp will be much less than the parts cost for the BA3 preamp.



Any comments/questions/suggestions are welcomed. Execution of this idea will have to await finishing the BA3 preamp and acquiring or re-purposing some enclosures.



Would you mind sharing a link for those buffer boards on eBay?
 
Calculate the output impedance of the preamp. How much capacitance can it drive without damaging the HF response? How many feet of interconnect cable is that? What happens when you connect a solid state power amp whose input impedance is 10K ohms, at the far end of a 5 meter interconnect cable?

Good questions. I will connect one of these up and see if I can device some torture tests.

I did perform an output impedance test a long time ago but have forgotten the results. I will re-do and report here.
 
You could easily convert that circuit to a B1 Rev. 2.
Q1 and Q4 can be omitted, just install jumpers between the emitter and collector of each of them. Omit also R1, R2, R7 and R8.

That is a great idea. I may buy some unpopulated kits and assemble them as a B1 R2. Thanks.

One attraction of the assembled boards is that they throw away 1/2 of the rail voltages. Great for a front end buffer driving a voltage gain stage. You can use the OPS rails.
 
Not autoformer...
Use of a level matching transformer as an autoformer can have performance advantages compared to the normal transformer mode with regards of parasitics and losses.
...You could call this an unbalanced to balanced converter with voltage gain. ...
For best result, an unbalanced to balanced transformer usually use a different winding technique and/or shielding compared to a regular level matching transformer.

I encourage you to take more measurements on the PC 600/15k and perhaps other alternatives since key performance factors will somewhat deviate from the M2 due to different mode of utilization. And for output, I woul think the Whammy output stage will serve you much better instead of the tiny buffers.
 
Use of a level matching transformer as an autoformer can have performance advantages compared to the normal transformer mode with regards of parasitics and losses.

For best result, an unbalanced to balanced transformer usually use a different winding technique and/or shielding compared to a regular level matching transformer.

I encourage you to take more measurements on the PC 600/15k and perhaps other alternatives since key performance factors will somewhat deviate from the M2 due to different mode of utilization. And for output, I woul think the Whammy output stage will serve you much better instead of the tiny buffers.


Hello Indra1. Nice to hear from you. Do you have a link to some technical info about these different transformer types?

I took your advice and had a look at the Whammy OPS. It looks like the same transistors from the BA3 preamp but configured as a follower. I have some of these and can give it a try to compare to the buffer boards.

I did order two other different configurations of the EDCOR line transformers to characterize.
 
Regarding the JFET input Cascoded Buffer Preamp Board use of Toshiba 2SK246/2SJ103, C2240/A970 699970145561 | eBay I would not know if the JFETs are actually 2SK246/2SJ103, but if one for one reason or another want to use 2SK170/2SJ74 instead that should be a piece of cake.

One attraction of the buffer board is that it is $11 populated for two channels. A quad of Toshiba 2SK170/2SJ74 from Punky Dawgs is $65 and then you have to add all of the other parts and PCBs. So,.. just investigating a lower cost possibility.
 
... Do you have a link to some technical info about these different transformer types?...
I have only scratched the surface on the subject, but it seems common knowledge that a phase inverted secondary is somewhat worse compared to in phase. As an example, Lundahl LL1660s employs a "internal Faraday shields to improve balance in phase splitting interstage applications" compared to the LL1660 which is the regular interstage/line output. Here in Papaland, you'd probably get way more from ZM, people used to call him Choky, or talk to tube forum guys deep into iron if you need technical details like 45, Bud Purvine and Wavebourn.
 
I have only scratched the surface on the subject, but it seems common knowledge that a phase inverted secondary is somewhat worse compared to in phase. As an example, Lundahl LL1660s employs a "internal Faraday shields to improve balance in phase splitting interstage applications" compared to the LL1660 which is the regular interstage/line output. Here in Papaland, you'd probably get way more from ZM, people used to call him Choky, or talk to tube forum guys deep into iron if you need technical details like 45, Bud Purvine and Wavebourn.

Thanks. That gives me some ideas. If there are some parasitic paths that cause imbalance in one transformer, it may be possible to use two separate transformers per channel and also add a mechanism for twiddling the balance between halves. That will add complication, parts and cost.

One of the ideas is to cobble together an instrumentation amp that can measure the goodness of the balance between halves.

cheers
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.