LA2a Gain control normal operating position

[Rob Flinn]

I have searched the forum for posts regarding what I am about to say & didn’t really find a solution/answer.  I may be raking up something that has been covered before.  If there is something to be read please point me in the right direction.  I have read about the 12AY& & 100k anode resisitor mod.
Anyhow possibly a silly mistake on my part, but here goes :
Over the years I have made a few La2a replicas, 2 that are point to point & 2 using the drip version 1 board.    Several people have used them over the years that are very experienced using different original vintage versions have said that for normal operation they have the gain control set to around 11-12 o’clock, whereas on all of my replicas it is more in the 8-9 o’clock range.  I have also been told that the Bloo versions behave like the original in this respect.  To be a bit more specific this is if I inject a +4dB signal into one of my La2a’s & switch the meter into the +4dB range, & then adjust the Gain control so I get a zero reading on the la2a meter.  THIS IS WITH NO COMPRESSION HAPPENING.  Can anyone verify how their original or replica behaves in this respect ?
Recently I had the opportunity to compare one of the drip versions (using Sowter 1009/4383 combination) against a vintage version & found that this discrepancy was indeed the case.  The vintage unit had a T4b with 3 ldr’s, but I have swapped  that with one of my UREI t4b’s & the response was the same.  These particular Sowters are supposed to be near replicas of the A10/A24 (& sound really good by the way).
All of my La 2a’s use real UREI t4b’s They use log pots for gain & peak reduction.  The components are the values as on the diagram labelled C-10953 C.  With regards to transformers I have several different combinations
In Out
Sowter 1009 Sowter 4383
Gardeners Sowter 8940
Parmeko Sowter 8940
UTC A10 Parmeko
It is fair to say that the versions with Gardeners & Parmeko input transformers have the Gain control slightly higher, but they have lower ratios, which would account for the difference.
Unlike most people I DON’T want to reduce the gain of the compressors particularly, so the 12AY7 & 100k anode resistors mod isn’t something that I want to do particularly.
I do like the sounds of all of these compressor, & they are all slightly different, & ultimately that is what is important, but I’m puzzled by this difference, & just wondered if anyone had a handle why I’m getting a difference, because I’m scratching my head now ?
Any thoughts would be appreciated

 

[hop.sing]

The only reason I can imagine for that behaviour are different log laws in the volume pots. Maybe you should measure the resistances created by the pots and see if they measure the same for the same volume but  simply at different positions.

Bbest Tobias

 

[Rob Flinn]

As far as I'm aware there are not different types of log pots ?

 

[Biasrocks]

Same issue here Rob, my LA2A A10/A24 version has way too much gain.

Perhaps the vintage versions that everyone are using have pads installed on the input
as detailed in the LA2A manual.

Mark

 

[Rob Flinn]

Hi Mark

Tanks for the response, I already thought about an input pad.  I looked at how much the signal would have to be reduced to make the pot line up the same as the vintage units & it was around -20dB.  The problem with this is that it also reduces the signal available to the sidechain amp, which, in turn, means that you can hardly get any compression happening.

Could you tell me where you gain pot is, it you set  your La2a up as in my 1st post ?

 

[hop.sing]

As far as I'm aware there are not different types of log pots ?

But somewhere I have read that there are different ways to achieve the log behaviour, and that today the pots are not made truly logarythmic anymore, because it would be to expensive. So, if everything else (transformers, circuit, t4b, no pad) is equal between the old and the new LA2as, there is not very much left to explain the difference but the pot itself.
Maybe it would be helpful to measure the absolute gain (Volumepot completly to the right) of the units, if it is the same on the old and the new ones, than it has to be the pot itself.

 

[Rob Flinn]

As far as I'm aware there are not different types of log pots ?

But somewhere I have read that there are different ways to achieve the log behaviour, and that today the pots are not made truly logarythmic anymore, because it would be to expensive. So, if everything else (transformers, circuit, t4b, no pad) is equal between the old and the new LA2as, there is not very much left to explain the difference but the pot itself.
Maybe it would be helpful to measure the absolute gain (Volumepot completly to the right) of the units, if it is the same on the old and the new ones, than it has to be the pot itself.

OK, However i just looked at the spec of the OMEG pots that I used in this position & the tolerence is within 10% of a log law, & the difference in the pot rotation is far more than 10%.

 

[jrowell]

I handwired a Bloo about 4 or so years ago.

I believe that came as a kit with most of the mechanical parts, including the gain and output controls.

The pots supplied where the PEC continuation of the original Allen Bradley pots.

I am right in the neighbourhood, so maybe when I am bored I will check into this. Allen Bradley Canada had a factory in Guelph, ON. I am pretty sure they made the classic AB pot (Type 'F' I think) there.

After AB got out of that line, I am thinking it was sold to PEC, who are in Toronto. They are basically identical (minor appearance differences)

They are also expensive. Digikey's $$ are now horrible on these (and a lot of other things)

For standard singles I can beat Digikey's price, fairly well, I think, and still make a bit of money for my effort.

I would have to gauge interest, and bring some in, the lead time is sort of long for one offs (3 wks)

But they feel, taper and set up in a Bloo LA2A exactly like the original.

The Bloo matched factory specs, and an original we ran it against in every way. Except the Bloo sounded much better. The orig was restored, but later set up, and with A10 + A24. Levels, specs and feel/action matched though.

We used the early component set up, and a vintage HA100 input (no triple shield to peel off that upper top)

I think the Bloo was the only kit to use these PEC's

Try them, and see what you think.

Although it sounds like this is a lot bigger difference then a log taper variation.

Essentially modern 'log' pots are closer to linear than the old ones, and come up a lot faster, then do progressively less, as you rotate.

The old style and the PEC come up somewhat slower.  

 

[Biasrocks]

Hi Mark

Tanks for the response, I already thought about an input pad.  I looked at how much the signal would have to be reduced to make the pot line up the same as the vintage units & it was around -20dB.   The problem with this is that it also reduces the signal available to the sidechain amp, which, in turn, means that you can hardly get any compression happening.

Could you tell me where you gain pot is, it you set  your La2a up as in my 1st post ?

Well, with a 0db signal I can manage to get the gain control up to around 10 o'clock, +4db signal
it's more like 8 o'clock. I usually can't advance my Peak Reduction control beyond 6-7 o'clock to get
the the kind of reduction I like to use, so the input pad would actually improve that situation and
get it into a more workable range.

I'm using PEC's in my build, but also before I got those I had ALPHA's in there and the behavior was
pretty much the same. I did by mistake have a 50K pot installed for PR and that seem to work a lot
better in terms of control range.

Mark

 

[emrr]

Another confirmation that pot taper can be different.  Manufacturers offer various tapers, but only a few make it to the retail market.  You see various tapers in Fender replacement volume pots for example, to give people a broader sweet spot in the throw. 

I am doubtful the 12AY7 mod affects gain at all.  The lesser tube gain will simply mean there's less in the NFB loop, and you end up with a looser more euphonic sound. 

 

[Rob Flinn]

I was just looking at the PEC pot on the digikey site & it seems that it also has a 10% tolerance.

I guess the point is 10% against what exact law !

 

[Biasrocks]

I just installed the suggested H pad into my LA2A. It makes the unit way less unruly and the gain/peak reduction controls seem to be in a much more reasonable range. It seems to toughen/tighten up the sound a bit as well, which is a good thing in my book.

It turns out to be a 10db pad on the input. I made mine switchable in case I want to return to full gain.

My suggestion Rob, is to install the pad and be happy.

Mark

 

[Rob Flinn]

Mark I'll give it a try.    The unit mine were compared to definitely didn't have a pad, because I checked that out the last time I was over there.

 

[lassoharp]

Space permitting you could try making a gain pot to your taper preference with a rotary switch.

Also the possibility of using a 600r ladder on the input though you'll have to take the insertion loss. Maybe the possibility of losing the 68K loading resistor there with permanent ladder termination, which will make up a small bit of the lost gain.

 

[Biasrocks]

From the LA2A operation manual

"For ease of control and to prevent overload of the input transformer, sufficient fixed pad
should be placed ahead of the LA-2A to allow normal output at approximately 50% setting
of the gain control."

Mark

 

[Rob Flinn]

From the LA2A operation manual

"For ease of control and to prevent overload of the input transformer, sufficient fixed pad
should be placed ahead of the LA-2A to allow normal output at approximately 50% setting
of the gain control."

Mark

Yes, I read that too, but the sufficient pad turned out to be  -20dB, which starves the sidechain of signal, like I said in a previous post.

 

[Kingston]

As far as I'm aware there are not different types of log pots ?

Indeed they all comply with the same log law. But they still vary wildly. The 10% tolerance alone is pretty bad, but remember it only applies to that particular manufacturer. And the tolerance is only measured on the end-to-end resistance. It does not take into account the log curve itself because this would be far too expensive to test. The curve in the carbon wafer can wildly vary ever which way.

Are there some high-end pot manufacturers with accuracy guarantees? I would certainly like to know.

Don't expect to match a Piher to Alpha to Omeg, ever.

I was just experimenting with this in several builds, one of them being LA2A. I have some Piher pots that actually matched really well, but the feel is just awful. Then I have some cheap Alphas with great feel, but these are absolutely impossible to match, especially the log curve. Equal gain so that one pot points at 10-a-clock, another 11 and a half, which is pretty unforgivable in LA2A where the first half of rotation should be extra accurately balanced due to the very high gain and the fact we only want the first 10-15dB or so of make up. Peak reduction pot is far less sensitive to this.

Oh, and there are three separate pot manufacturers named alpha, which doesn't exactly make this easier.

The only way to match them perfectly and with flexibility is a 24-step rotary switch.

I don't like the input pad or low ratio transformer solutions either. LA2A sometimes wants the needle buried, and other times we might want to use it as a preamp.

 

[Biasrocks]

I don't like the input pad or low ratio transformer solutions either. LA2A sometimes wants the needle buried, and other times we might want to use it as a preamp.

That's why I put it on a toggle.

20db seems like way too much Rob. My pad ended up being 10db using the 500 ohm tap on the A10, which was just enough to put it into an optimal range. I think the limiter sounds better with the pad engaged but I think it has more to do with getting the output stage running at slightly higher gains.

Mark

 

[hop.sing]

I don't like the input pad or low ratio transformer solutions either. LA2A sometimes wants the needle buried, and other times we might want to use it as a preamp.

That's why I put it on a toggle.

20db seems like way too much Rob. My pad ended up being 10db using the 500 ohm tap on the A10, which was just enough to put it into an optimal range. I think the limiter sounds better with the pad engaged but I think it has more to do with getting the output stage running at slightly higher gains.

Mark

Since the volume pot on the la2a is simply an attenuator and not a gain control it is unlikly that you like it better because of the output stage working harder. It is fixed gain. But the input transformer gets less signal and distorts less and that makes a difference for sure.

 

[pacemaker]

Hi Guys,
and was interetsed in some PEC pots,
but i saw this thread :
http://www.groupdiy.com/index.php?topic=40948.msg507957#msg507957
and thy didn't seem to be as high quality as announced,
any feedbacks on these,
Biasrock, did you used the pad schematics from the LA2A manual ?
and Rob, wht's your impression with the sowter 1009 ?
regards,
francois