LA-2A point to point build Sowter TXF

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Hi,

AC wires all twisted nice and tight, grounding scheme may need looking at as have tried to copy the original with the the isolated pins to group up all connections close to the tubes.
From the IEC I feed a gnd to the xlr's and another feed to gnd B+ and H+, there is another point on pin6 on 6QA5 where the 510pf goes to a solder tag on the chassis close to the tube.
Is there a specific plan I should implement here, happy to do this as specific as possible.

Regards

Spence.
 
Hi,

Have had another look at the gnd scheme, it seems the gnd scheme is split into 4 point sections, and I can see from the original that the power TXF are grounded to the IEC gnd along with the 47uf and 33uf, nothing else is grounded at this point.
These next ones need your solder tag idea I think? Rather than the isolated pins which then feed all the way back to the IEC with long wire.
The next is by the T4B which had the input TXF, 1k V2, 1.5k, 2.7k and 470k.
3rd section is gnd of input XLR pin 1, 47k/330pf, 510pf, 100uf/1meg.
4th is output TXF, 470pf/ Zener, 330k.

My gnd scheme is not like this so I'm going to change it and see if it improves.

What do you think?

Regards

Spence.
 
hi,

have just redone the gnd scheme as ive detailed above and now the unit has no hum on output!! hooray!!
still got the acoustic noise from the case of the txf but this will be sorted once i install the toroidal hopefully.
very clean punchy sound, lacks a bit of bottom end but that is OEP txf, once i get the Sowters in there full bandwidth will commence but i am predominately using this for vocals so a natural roll off on the low end is good.
not a bad unit for £255!!! very happy me!!

regards

Spence.
 
Hi,

Starting on 2nd unit now, have all bits together, will add in jack socket for stereo connection on both units.
Will put up pictures, hopefully this build will go better than how the first went!

Regards

Spence.
 
Hi,

Have just wired up the power toroidals and checked voltages, getting 250v and 2 x 10v AC.
Have then put the 250v to the bridge rectifier and when unit is not turned on at power switch on unit but IEC has power I get 5v at the bridge rectifier? Voltage is not 350v as it should be at bridge as well?
Only 275v at + at bridge?
Something is a miss, need to sort this.

Regards

Spence.
 
Hi,

my second version is now sort of working, getting good voltages throughout and have signal to output, just dont have compression yet? i know the T4B works as i have tried the other one from my working unit. must be something simple hopefully.
Also the gain is very hot, with the gain hardly turned up there is loads of signal, goes into distortion very quickly?


regards

Spence.
 
hi,

gonna get all my valves tested and put in the best pair as the ones i got in there are defo shot to pieces! a/b with a stereo output from a dat machine going into the unit for the left and right hand channel into desk, setting levels and then switching between them reveals a top end loss.
also going to think about replacing the plate resistor of the first 12ax7 from 220k to 100k and replacing the valve for a 12at or 12au7.

regards

Spence.
 
Hi,

Have  second unit all working now, although a cold solder joint was playing up.
Good compression and output just need to fine tune both units.
I have put in 12AT7 and they sound good, Going to test with 12AU7 as well and check biasing.

Regards

Spence.
 
Hi,

Have put in 12AU7 into first unit and compared with second unit having 12AT7 both are mullards.
First unit to my ears sounds slightly better.
I have not changed the anode resistor yet as I want to have a bit more technical data for changing the 220k for 100k.
Could anyone explain what is happening with the change please.
Also do I change both 12AX7's or just V1
And do I change all anode 220k as there are 3 of them R9, R33 and R13.

Clarity on this would be excellent please, but laymens terms preferred.

Regards

Spence.
 
Hi all,

Just thought I'd report on my findings now I've had a chance to put both units through their paces now.
Both units compress a lot and I can get a very high level of audio down when using them with Neve and SSL9k pre amps.
If anything I will investigate being able to bring the amount of compression down a little.
Also the amount of distortion is in question and would like to get both units a bit better in this department.
One it is actually a little better, so with some testing now I will find out why.
I'm sure the distortion will be reduced once I use the sowter TXF.
If anyone has any ideas on either of these fronts please chirp in.
Other than these points which to be honest with setting the gain and peak reduction very low I can get a very good sound out of these units.

Regards

Spence.
 
Spence,

First of all, well done for getting this far!

Changing the first tube from a 12AX7 to a 12AT7 or 12AU7 without adjusting the relevant resistors will screw up the working points that have been so carefully chosen by the original designer.  Compare the triode curves of these tubes (Franks site) http://frank.pocnet.net/  and draw in the load lines for 220k and 100k resistors and see what happens to the voltages.

There is a DC feedback circuit running to V1 so if its cathode voltage changes, so will the feedback.  This is important to maintain stability and reduce distortion, noise and output impedance, keep the same current through it whatever else you do.

In general, both 12AT7 and 12AU7 will draw more current so by ohms law the voltages across the resistors will change.
You could carefully reduce the resistors to keep the same operating points, gain of 12AX7 will be say 55, 12AT7 say 35, 12AU7 would be around 15.  Personally I would use a lower ratio IPT to reduce the gain, this will have lower frequency losses as a bonus and probably have less tendency to ring as the inductance will be lower.

You don't need to touch R33, or change the tube as that will drastically reduce the drive to your TB4's.

Check the spec of your OPT to see if it can handle the output of the LA2A, the original could give up to 16dBm into 600 ohms.

best
DaveP

 
HI Dave,

thats very interesting what your saying as my results have shown that both my LA2A builds very quickly go into distortion with the ECC81 in them.
Loads of compression on both units but one of the units has slightly less distortion, i say distortion but it sound a very hot signal.
i will try changing the 220K down to 100k, just to make sure im changing the correct one, do i just change R9 in V1 or do i change both R9 and R13?
both my units do sound great but now i have given them a bit more of a listen i could make them sound much better.

regards

Spence.
 
Hi Spence,

The original schematic has V1 with 1.05V on the cathode and V2 with 1.1V and you should aim to keep somewhere in this ballpark, certainly not less than this.

There is a well known error on the schematic regarding the voltage at the junction of R11 and R21, the schematic says 27V but it is in fact 19V.  If you calculate the current drops through resistors R10, R11, R21 & R22 you will see that 19V is the only possible voltage when the mid point of the 12BH7 is 100V (which it has to be for correct operation).

This means 0.26mA of DC is flowing into the cathode of V1 so you must arrange the resistor values to maintain this current.
A quick look at the plate curves shows that to maintain the same voltages on V1 with a 12AT7 you will need to change R9  to about 27K and R10 to about 270, these are ballpark figures.  This is because almost 10x more current will flow.  The HT dropper R16 will have to be reduced in order to maintain around 200V for V1 and V2, this is because much more current must flow through it.

If you lower R9 then C1 and possibly C2 may need increasing or you may lose bass response.

This process shows you what it is like to design an amp and the thought that went into designing it, Ohms law is your greatest friend.  Your 12AT7 is going into distortion because the operating point with the original resistors is almost at 0V on the cathodes so you have no headroom at all.  A little more understanding of how your amp works will increase the pleasure you get from it and help you iron out its faults

best
DaveP
 
Hi Dave,

You are a fountain of knowledge on this and this is very helpful information.
I have changed R9 to 100k and no noticeable difference yet in either unit.
I need to take this latest information and check voltages and implement what you have said.
Need to find out why my second unit is sounding a bit more driven that the first, both have same values and identical parts.
Maybe a dumb newb question but does the limiter response control pot have much influence on the overall compression?
Both units seem to sound pretty good when I use my api-312 preamps and Ssl preamps but when I use my NEVE 1290 it sounds very distorted?

Regards

Spence,
 
Hi Spence,

Measure the voltages you are getting for the two units and post them here so we can see what's going on.

If you can, measure the ac input voltages from your pre amps, maybe the Neve is hotter, but we can't tell until we see some figures.

best
DaveP
 
Hi DaveP,

Your my new hero!!! Thanks for all this wonderful advice, I do have a few little problems with my diy units that I am trying to sort so I have trouble free recording, and unlike you I'm still very much a beginner in the world of electronics.

Voltages on la2a are as follows :

First unit

V1
Pin1 = 94.7v
Pin3 = 1.478v
Pin6 = 70.9v
Pin8 = 1.238v

V2

Pin 6 = 212v

Second unit

V1
Pin1 = 102v
Pin3 = 1.315v
Pin6 = 81v
Pin8 = 1.067v

V2
Pin6 = 202.2v

All voltages seem a fair bit off?

What would you advise.

Regards

Spence.
 
Spence,

I'm no hero, I just did a lot of research on this comp before I made it (mine uses green LED's instead of the TB units)

As beginners go, you have done very well to make two with so little hassle.

Those voltages are all within the normal spread of tube parameters.  Your tubes have plenty of gm (current draw) as evidenced by the low anode voltages and healthy cathode voltages.

The output tube is set up as a "White Cathode Follower" (WCF), this has to be in perfect balance or there will be distortion.  The minimum distortion point is luckily the same point as maximum output, so you can adjust R17 to achieve this (use a 22k pot).  Be careful when you are working on it live!  This may be the source of your problem, but if it ain't, you will need to look elsewhere.  This is where a scope would be handy, you can follow the trace through the amp until you find the problem.

Your generation has probably got an app for a scope on your phone!

best
DaveP
 

Latest posts

Back
Top