4343 - L pads removed, good result

[Goldjazz]

A friend recommended I take the Lpads out of the signal path and replace them with fixed resistors. Basically I removed them and measure the resistance of them at the setting I had chosen previously. Then replaced them for fixed resistors at that resistance. I didn't use an fancy resistors. I has previously cleaned the lpads with deoxit and didn't think much of them as a source of signal degradation. When I popped a couple of then open they didn't appear to be in such bad shape. But anyway I proceeded with the job.

So the result has been actually quite amazing. Much livelier, clearer and more sensitive for all drivers the Lpads were connected to. A lot more sparkle inthe uhf, more startling clarity form the hf horns, and more punch from the MF. I'd actually say this was a bigger improvement that getting rid of that awful biamp rotary switch.

Next step I'll try some fancier metal oxide etc resistors. Also I'll take some freq response measurments to check if the response is flat, but sounds pretty flat to my ears.

Highly recommend trying this for any older Studio monitors with Lpads. Obviously you lose the ability to tweak values on the fly. I do wonder if high quality new lpads would be the answer or simple if all those sliding contacts are just a bad idea even with fresh ones. It was a little bit of work, but well worth it, definetley not going back.

So here's where I'm up to with the tweaking of my 4343s

- Biamp switch removed: Say a 10% improvement. Better clarity.
- Baimping with a First Watt F4, again a 10% improvement over not biamping or biamping with a poor crossover. Better bass definition
- Getting really solid stands that put the monitors to a height such that the bottom of the hf lens is at ear level. 5% Better Bass.
- Getting rid of Lpads: 20% improvement. Better clarity, liveliness

Next Steps:

- measuring Freq response, tweaking resistance values to get flattest response, and replacing resitors with good ones.
- Rebuild whole network.

[Ricpan]

Thinking about doing this myself on my 4345s. Did you locate the resisters near the old L pads? I also may rewire the speakers with some Dueland 16ga tinned oiled cotton wire while Im in there.

[Ian Mackenzie]

Great thread.

I was actually measuring a 4345 today looking at the L pads full and "0" with Linearx LMS

The hard part is getting a clean accurate measurement.

But what the L pad does is not linear with frequency or level.

I plan to map the incremental response curves from +3 to "0" to -3 Db for the mid horn and slot and then code a fixed resister network with a quality switch

It might upset the local natives and pets as it needs to be done outside......Lol

It's going to take a while ...

[Ricpan]

Ian thats the way my supertweeters are set up with my Bozak Concert Grands. I guess just run a cd of pink noise and get the outputs close with a sound meter and take an Ohm reading across the lpad and bypass the Lpad install the resistor and thats it. Im I on base with this though?

[Goldjazz]

Thinking about doing this myself on my 4345s. Did you locate the resisters near the old L pads? I also may rewire the speakers with some Dueland 16ga tinned oiled cotton wire while Im in there.

I simply remove the Lpads and measured the resistance they were delivering and made a separate mdf board with all the resistors mounted. I have since checked the frequency response, and it's pretty equivalent to what it was with the Lpads installed.

I'm not sure what the improvement is due to. As I said the Lpads seemed reasonably clean, perhaps there was some corrosion there somewhere, or simply having all those contacts is just not a good idea for signal to pass trough.

Yeah requiring the speakers is also something I'd like to try. I think these beauties are worthy of much better wiring caps etc. One thing that's always impressed me with the 4343 is it's ability to keep deliverying improvements with increimental tweaks. I feel there's still quite a bit left to get out of them.

[Goldjazz]

Great thread.

I was actually measuring a 4345 today looking at the L pads full and "0" with Linearx LMS

The hard part is getting a clean accurate measurement.

But what the L pad does is not linear with frequency or level.

I plan to map the incremental response curves from +3 to "0" to -3 Db for the mid horn and slot and then code a fixed resister network with a quality switch

It might upset the local natives and pets as it needs to be done outside......Lol

It's going to take a while ...


Very interesting, would love to see those results. This was a quick and dirty experiement to see if there was a benefit, so I was just using $1.80 jaycar resistor. So next I had planned to rebuild the resistor network with quality resistors and in a way where I could swap out different values easily. Then use the old lpad as a means to tell me what resistance values to try.

[Ian Mackenzie]

Very interesting, would love to see those results. This was a quick and dirty experiement to see if there was a benefit, so I was just using $1.80 jaycar resistor. So next I had planned to rebuild the resistor network with quality resistors and in a way where I could swap out different values easily. Then use the old lpad as a means to tell me what resistance values to try.

That sounds like a nice diy project

The fact that you are trying out the idea is very cool.

Let me know how it goes?

Somewhere along the line "flat" as in measurement needs to be done overall so you have a realative zero baseline to work with and you are on to that

Once flat is established we can then decide on what degree of variation you need "up" and "down"

There are several approaches and possibilities.

As hinted previously l am writing series of new articles for a web page that will be very beneficial to the wonderful owners of thes classic systems (4343-4345)

I have always said they benefit from some rather insightful upgrades and fine tuning.

[toddalin]

I simply remove the Lpads and measured the resistance they were delivering and made a separate mdf board with all the resistors mounted. I have since checked the frequency response, and it's pretty equivalent to what it was with the Lpads installed.

I'm not sure what the improvement is due to. As I said the Lpads seemed reasonably clean, perhaps there was some corrosion there somewhere, or simply having all those contacts is just not a good idea for signal to pass trough.

Yeah requiring the speakers is also something I'd like to try. I think these beauties are worthy of much better wiring caps etc. One thing that's always impressed me with the 4343 is it's ability to keep deliverying improvements with increimental tweaks. I feel there's still quite a bit left to get out of them.

It is possible that the wiper on the L-pad provides a poor(er) contact, and certainly poorer than a soldered junction. When you measure the resistance across the wiper, with no load, it is nil. But when you put a load on the circuit and it has to pass some real current, it acts as a resistor impeeding the flow of current.

When I run my garden trains, I have stranded copper wire that "jumpers" the rail joiners to assure good contact across the joiners. Over time these jumpers can break down due to corrosion (20 years outside getting watered twice every day) and the trains will run slow/stall past a bad jumper. But if you put a voltmeter on that section of track, it reads the same as a "good" section, until you put a load on it. The trains use about the same amount of power as a speaker does.

[ivica]

I simply remove the Lpads and measured the resistance they were delivering and made a separate mdf board with all the resistors mounted. I have since checked the frequency response, and it's pretty equivalent to what it was with the Lpads installed.

I'm not sure what the improvement is due to. As I said the Lpads seemed reasonably clean, perhaps there was some corrosion there somewhere, or simply having all those contacts is just not a good idea for signal to pass trough.

Yeah requiring the speakers is also something I'd like to try. I think these beauties are worthy of much better wiring caps etc. One thing that's always impressed me with the 4343 is it's ability to keep deliverying improvements with increimental tweaks. I feel there's still quite a bit left to get out of them.

Hi Goldjazz,
May be such can help

https://www.don-audio.com/Rotary-Switch-2024-2-Pole-24-Position


regards
ivica

[Goldjazz]

It is possible that the wiper on the L-pad provides a poor(er) contact, and certainly poorer than a soldered junction. When you measure the resistance across the wiper, with no load, it is nil. But when you put a load on the circuit and it has to pass some real current, it acts as a resistor impeeding the flow of current.

When I run my garden trains, I have stranded copper wire that "jumpers" the rail joiners to assure good contact across the joiners. Over time these jumpers can break down due to corrosion (20 years outside getting watered twice every day) and the trains will run slow/stall past a bad jumper. But if you put a voltmeter on that section of track, it reads the same as a "good" section, until you put a load on it. The trains use about the same amount of power as a speaker does.



Interesting

Yeah I suspect something like that may be the issue. On several occasions in the past, in audio, I've been surprised by the beneficial effects of a good solid solder joint over mechanical contacts. Perhaps you're right, perhaps the bet way to asses a joint is to apply voltages or currents similar to maximum use case and see if the resistance has gone up linearly as expected. I'm not sure how a digital multimeter works, perhaps it just apply a tiny voltage and measures the current then deduce the resistance? I guess it would have to be a tiny voltage otherwise you could damage cirucits you are measuring.

[Goldjazz]

Hi Goldjazz,
May be such can help

https://www.don-audio.com/Rotary-Switch-2024-2-Pole-24-Position


regards
ivica


Thanks

I suspect this would provide a better contact than a sliding resistor. I do wonder if there is any difference between a mechanical contact such as this and a solder joint. Perhaps I'll get this and find out.

[Goldjazz]

That sounds like a nice diy project

The fact that you are trying out the idea is very cool.

Let me know how it goes?

Somewhere along the line "flat" as in measurement needs to be done overall so you have a realative zero baseline to work with and you are on to that

Once flat is established we can then decide on what degree of variation you need "up" and "down"

There are several approaches and possibilities.

As hinted previously l am writing series of new articles for a web page that will be very beneficial to the wonderful owners of thes classic systems (4343-4345)

I have always said they benefit from some rather insightful upgrades and fine tuning.


Excellent

Please send us a link when you're finished. I'm sure there are many like myself who'd love to read it.

[Ricpan]

What do you think of this idea? I have an Electrovoice DX38 processor. I can adjust the Db to the woofer and the Db to the high (Mid, tweeter and super tweeter). Have a fix resistor for the woofer and for the Mid that I adjust with the processor and use a L-pad for the tweeters.

[Dr.db]

Great thread.

I was actually measuring a 4345 today looking at the L pads full and "0" with Linearx LMS

The hard part is getting a clean accurate measurement.

But what the L pad does is not linear with frequency or level.

I plan to map the incremental response curves from +3 to "0" to -3 Db for the mid horn and slot and then code a fixed resister network with a quality switch

It might upset the local natives and pets as it needs to be done outside......Lol

It's going to take a while ...


Hi Ian,

have you proceeded on this project?
I would be very interested in an quality alternative for the l-pads.

Regards,
Olaf

[Ian Mackenzie]

It’s in the works

[Dr.db]

Great, I´m really looking forward to your results!