No doubt it is worth it to keep good things running. I agree. Especially cars! I cherish my 6 speed manual transmission as Corp America has all but eliminated it from every car to streamline production and ensure the highest profits. If my amp stopped working, I would pay up to get it repaired. But in my case, everything works and I enjoy what I hear. So paying $1200 (and probably more) to have it overhauled is not on my list of things to spend $$$$ on in this now very expensive world we now live in. I once almost bought an amp off Audiogon and when I asked the seller why he was selling, he told me he had taken his 25 year old amp to have it overhauled with all new caps and resistors and had all the contacts re-soldered. He said he didn't like the sound of the amp when he got it back. Many times I have had things repaired (not just audio gear) and didn't like how it performed when I got it back. So now that I am close to retirement age, I don't change anything in life that is working and I am happy with it. There is no "Spot on", everyone's situation is different. If you like what you have, leave it alone and enjoy the music.
Panel life?
- Thread starter zinda
- Start date
Help Support MartinLogan Audio Owners Forum:
All of this equipment we use is machinery and it wears out with use there is no disputing that fact no matter how you slice it and as it wears out the performance degrades that’s all I’m saying. Enjoy your equipment.
This thread has morphed from 'panel life' to 'equipment life' it seems, but it's all good stuff.
I too am a proponent of keeping old amps alive and well, even if they aren't the expensive 'high-end' models.
I'm sharing my recapping story below so that others will know that even someone with no electronics experience can recap an amp, if you're careful, without breaking the bank. Below is a link to my recapping thread, with photos, on the CarverSite:
https://thecarversite.com/topic/1116-recapping-a-tfm-25-for-dummies/#comment-7094
When building my system in 2008, I purchased three used Carver TFM-25's on Ebay for around $900 (total), and refurbished them. As I recall, I spent about $120 per amp on parts.
Even though I had no electronics experience and no clue how to refurbish an amp, I could solder pretty well, and I found service manuals and inspiration on the CarverSite Forum, and dived in. Fortunately all three amps were still in working order so all I needed to do was replace the components most prone to fail with age.
In each amp I replaced all 39 electrolytic caps, checked resistors, and replaced any that looked even a little toasty (even if they measured good).
A typical refurb would not involve raplacing any transitors, but in my case there was a TMF-25-2 Service Bulletin reliability upgrade, which I did concurrently, and it consisted of replacing (4) pre-driver transistors, replacing several resistors with 'flame proof' versions, and lofting all of the replaced components 3/8" above the board to prevent heat-damage to the board.
Before starting, I drained power from all the caps using a 100 watt light bulb. The big power caps lit the bulb brightly for a few seconds before dimming down as their charge drained.
Before removing any caps I marked the tops of every cap (with a Magic marker) so that I would not get confused about which caps I had already replaced and those I had not yet replaced.
Throughout the refurb I had the amp connected to speakers, and I would replace four or five components, and then do a sound check with music. That way; if something went wrong I wouldn't have far to backtrack to find the problem.
After each sound check, I drained power from all the caps again before continuing.
The last step was setting the idle currents to bias the transistors on each channel. There was an easy procedure for this in the Service Manual, which involved conncting DVM leads to check points on the board, turning the amp on and letting it warm up for five minutes, then using a tiny screw driver to adjust bias pots unitl the DVM reads 1.0 mOhms.
That's all there was to refurbishing a Carver TFM-25.
I found the experience really empowering. All three amps played wonderfully, and are still playing wonderfully to this day.
To my ears; the amps sounded OK before the recap but definitely more dynamic and alive after the recap.
Hey, go for it!
I too am a proponent of keeping old amps alive and well, even if they aren't the expensive 'high-end' models.
I'm sharing my recapping story below so that others will know that even someone with no electronics experience can recap an amp, if you're careful, without breaking the bank. Below is a link to my recapping thread, with photos, on the CarverSite:
https://thecarversite.com/topic/1116-recapping-a-tfm-25-for-dummies/#comment-7094
When building my system in 2008, I purchased three used Carver TFM-25's on Ebay for around $900 (total), and refurbished them. As I recall, I spent about $120 per amp on parts.
Even though I had no electronics experience and no clue how to refurbish an amp, I could solder pretty well, and I found service manuals and inspiration on the CarverSite Forum, and dived in. Fortunately all three amps were still in working order so all I needed to do was replace the components most prone to fail with age.
In each amp I replaced all 39 electrolytic caps, checked resistors, and replaced any that looked even a little toasty (even if they measured good).
A typical refurb would not involve raplacing any transitors, but in my case there was a TMF-25-2 Service Bulletin reliability upgrade, which I did concurrently, and it consisted of replacing (4) pre-driver transistors, replacing several resistors with 'flame proof' versions, and lofting all of the replaced components 3/8" above the board to prevent heat-damage to the board.
Before starting, I drained power from all the caps using a 100 watt light bulb. The big power caps lit the bulb brightly for a few seconds before dimming down as their charge drained.
Before removing any caps I marked the tops of every cap (with a Magic marker) so that I would not get confused about which caps I had already replaced and those I had not yet replaced.
Throughout the refurb I had the amp connected to speakers, and I would replace four or five components, and then do a sound check with music. That way; if something went wrong I wouldn't have far to backtrack to find the problem.
After each sound check, I drained power from all the caps again before continuing.
The last step was setting the idle currents to bias the transistors on each channel. There was an easy procedure for this in the Service Manual, which involved conncting DVM leads to check points on the board, turning the amp on and letting it warm up for five minutes, then using a tiny screw driver to adjust bias pots unitl the DVM reads 1.0 mOhms.
That's all there was to refurbishing a Carver TFM-25.
I found the experience really empowering. All three amps played wonderfully, and are still playing wonderfully to this day.
To my ears; the amps sounded OK before the recap but definitely more dynamic and alive after the recap.
Hey, go for it!
Last edited:
Yeah, if you're happy with it and it's running fine I probably wouldn't do a thing.
BobLobLaw
Member
The main (but no only) factors responsible for panel degradation are moisture, UV exposure and environmental contamination (dust, smoke, etc).
Moisture is by far the biggest enemy and will degrade the panel faster than anything else. An air conditioned room is about all you need to prevent moisture degradation.
UV light degrades the vapor deposited coating but glass just so happens to filter out the UV responsible for the degradation. So in front of a window is fine as long as the sun filters through glass first. No direct sun contact.
When the panels are built, the film gets coated by hand with a sponge dipped in a mixture of graphite and alcohol that then gets spread onto the surface of the film. This helps prolong the life of the film by adding a 2nd coating to the already vapor deposited PET film (Polyethylene terephthalate)
The symptoms of environmental degradation, (usually dust) will be a crackling sound from the panel when powered on. Dust carries a static charge and is naturally attacted to the positively charged film when it's in use, eventually becoming a pathway to ground causing the crackling sound. This was minimized years ago when ML started using a sensing power supply that allows the panel to discharge and power down when not in use, therefore losing its static attraction to dust. Vacuum both the front and back of the panel with a strong vacuum to get rid of most crackling.
The symptoms of degradation due to moisture will simply be a loss of volume output.
Rarely do both panels degrade at exactly the same rate so one panel will usually start to sound quiter than the other when they need replaced.
Moisture is by far the biggest enemy and will degrade the panel faster than anything else. An air conditioned room is about all you need to prevent moisture degradation.
UV light degrades the vapor deposited coating but glass just so happens to filter out the UV responsible for the degradation. So in front of a window is fine as long as the sun filters through glass first. No direct sun contact.
When the panels are built, the film gets coated by hand with a sponge dipped in a mixture of graphite and alcohol that then gets spread onto the surface of the film. This helps prolong the life of the film by adding a 2nd coating to the already vapor deposited PET film (Polyethylene terephthalate)
The symptoms of environmental degradation, (usually dust) will be a crackling sound from the panel when powered on. Dust carries a static charge and is naturally attacted to the positively charged film when it's in use, eventually becoming a pathway to ground causing the crackling sound. This was minimized years ago when ML started using a sensing power supply that allows the panel to discharge and power down when not in use, therefore losing its static attraction to dust. Vacuum both the front and back of the panel with a strong vacuum to get rid of most crackling.
The symptoms of degradation due to moisture will simply be a loss of volume output.
Rarely do both panels degrade at exactly the same rate so one panel will usually start to sound quiter than the other when they need replaced.
I had heard that a secondary graphite slurry coating was applied and you just confirmed it.
Now I have a question:
My one and only experience with ML panels was replacing the diaphragm in an older ML Theater panel. The diaphragm in that panel had a lighter coloration tint in the areas over the spars and about 3/16" on either side. I concluded that those areas over the spars had been masked off before the slurry coating was applied.
I speculated that these areas were masked off either to facilitate better bonding to the mating spars (on the front stator), or as a barrier against charge migrating up the spar edges to the front stator, as might occur if dust builds up along the spar edges. Just a theory...
Do you happen to know if those areas were masked off, and if so, for what purpose?
BobLobLaw
Member
The film is masked around the edges and over every spar, prior to adding the graphite coating to keep the coating away from the edges and off the spars. When the masking is removed it also removes the vapor deposited coating effectively turning the panel from 1 large speaker into several small individual cells or windows.
The edge of the panel is where the powder coat is the thinnest so removing the vapor deposit from the edge of the film prevents the voltage from the film from ever arcing over to the edge of the panel should the powder coat ever be too thin and not caught during QC.
That makes sense then, as the diaphragm-to-stator spacing is closest over the spars, and charge tends to migrate to areas closest to a stator.
Thanks for the info!
Really fantastic info, new member from Lawrence, Kansas
Tosh
Well-known member
Here's a pic showing the graphite between the masked-off areas. I didn't realize the vapor coating is removed during the assembly...
Attachments
BobLobLaw
Member
Yes you can definitely see the "windows" left over on the film. The panel builder here is getting ready to apply the graphite bias strip (tape) by first attaching it to the bare part of the red wire, then running it halfway across the bottom, turning it, and running it up the entire side of the film, all in one continuous piece.
Russr
Well-known member
Hey, BLL! Almost sounds like you were there...!!!???
BobLobLaw
Member
Built stat panels for several years.
I know you have a good grasp of ESL panels and maybe you can shed light on why JansZen designed panels don't seem to be affected by age. I used 16 panels from 1959, 4 form 1969 and 8 from 1971. They all had the same output. The 59s were true JansZen and the others were RTR made.
JansZen used a heavy paper the covered the backside on their 5"x5" sized, they also used them with sealed back enclosures, I thought this was a bad idea since they worked so well when removed from their cabinets and mounted in open air with rear reflections doing their magic. The 71s were Infinity and they are 2.5"x 5" and no backing and were mounted with the backs open to reflect. They also suffered from poor design in how the cabinets were made. For some reason, they failed to make mirrored sets of cabinets. The panels had the back open but they had to go the full depth of the cabinet before they hit open air.
They angled the upper (when stood up like a normal speaker) panel upwards at a 12 degree angle. If they were laid in their sides the angled panel would be the left side panel on both unless the speaker was flipped over . Infinity also used a terrible 12" woofers along with an unneeded Philips duocone (whizzer) midrange. They sounded awful once again do to poor design and use of what was otherwise a very good driver.
Why was it so difficult for makers to come up with the proper way to maximize their output and realize the full potential? You'd think with a small amount of thought common sense would dictate that they had much more potential than what they were doing. I can confirm this by remounting them and finding a much better woofer integration.
I have now spent 2 months with the Aerius and have had time to try and find out what design flaws they also have.
First off, I can't get the same depth and full sound from them as I could with JansZens and it was simple to do with JansZen panels. No real trucks, 1 hour of set up and it was like magic with a perfect sound that floated in front of you while sounding like huge headphones.
I came to a conclusion that the curving of the panel is causing the rear waves to reflect and destroys the otherwise perfect straight line path to the rear walls so the signal can travel back to the front and reinforce the front waves. I know this simply by tests I did while designing my repurposed panels configuration.
I first thought that curving them (as JansZen did with model 30 and 130 cabinets) I could create a larger sweat spot. I found that was not the case, what happened was the sweet spot vanished and they could not be positioned in any way that could restore the effect I had so easily found with just panels per side in a flat plane.
I could never get the model 30s to provide that same effect and realized that they needed to be changed and placed in a flat plane that allowed a path to the rear wall to once again make that magic stage reappear. After trying them in many configurations, I used thin bent metal strips that I mounted the on and then bent the metal to hold them in different angles.
The best way I found was to mount 3 panels in a flat horizontal plane with a 4th panel mounted on the outer edge aimed slightly towards the side wall. The other 3 were aimed nearly straight forward.
I did this with 2 groups above and below a 10" woofer then added more panels in either side of the woofer like a sunflower.
This eventually was adjusted to place all panels in the same plane and positioned about 1" back from the face of the 10" driver.
I did not give up on my efforts to get the Aerius to achieve the magic effect. Even though my first thought was to cover the rear waves and stop the colorizing and diminishing reflection of the rear waves. I put some thought into reinforcing the back wave by adding a solid wave from a second woofer that would bring some coherent binding force, even if it was combining with ⅓rd if the rear waves properly, the full low frequency wave could make a huge difference on e it reflected to the front waves.
So made some test rear panels and removed the crossover since I'm active and wouldn't use the any way. I found 2 cooked resistors on the High voltage board, both were still functioning, due to the way they were mounted, they had burned the board pretty crispy. So rather than remove them I cut the body out and soldered in new ones that are off the board in height, I soldered onto the remaining wires so I didn't disturb the copper traces on the crispy board.
I also noticed that the connection points on the transformer where the wires connect via tiny screw terminals were all quite loose.
I would suggest people check those connections if they have issues. Another problem I found was the brown wire from the transformer I one board was detached. It most likely came loose while doing work since it should be soldered on. After close inspection it appears that the wire may have had a bad strip job that ended up removing half the wires and left only a few strands to work with.
I stripped it back using a lighter and compared what I saw left behind in the solder joint to confirm. I sealed the "sealed" cabinet, it's not even close to be considered sealed unless, sealed means no port? Hot glue in every place possible including the spike holes and the backside of any screws that penetrated the motor board. I alwAys try to use screws that don't break the surface I the backside of the motor board when I'm making a cabinet. Sometimes I'll glue pieces of wood behind each screw hole on speakers to create a sealed chamber and added more secure mounting.
I positioned the woofer up high near the panel waves as possible to help it reinforce the rear waves. I also have the panels parallel with the rare wall. I feel that MLs tipping if the panels is another mistake, the rear waves are then aimed upward and create further reflections and mistiming with front waves.
It's almost like ML knew how to make a panel driver but that's it, they failed to examine the entire picture that includes exactly what's going on with the sound waves and what's most important to get the most from them.
What I'm talking about should automatically be obvious and common sense when developing a proper design. I know they are concerned about appearance, but in my book, that's secondary to function. You work with what you have once you've maximized the potential.
I tested the panels output in full range tests to find the best place for the crossover point. I saw useable frequency down to 250Hz but there's a dip after that and at 350Hz it's back up to nearly flat with gain around 800Hz to 1300Hz which can be corrected easily with an EQ.
There's also a chance that it may e smoothed out on e the wooofers are connected. I still had to test the woofer output as a constant and reference point. I changed out the stock woofers as well. So I tested just one woofer in one cabinet. Then after I added the second, I tested it in 4 more tests that involved polarity or the wires and the way the driver was mounted.
Then looked at the plot to see if there was anything that jumped out at me to be the ideal crossover point. In this case, plot that has the woofer with a dip around 250 and a peak after then a dip would have been ideal and would minimize the crossover slope to help transitioning between drivers.
Knowing the timbre of the woofers being much different than the panels especially at 250Hz would make the transitioning obvious, so working with a mix at the crossover point is a must to integrate the sound properly .
I got kind of lucky and found a dip at 350Hz that matched the slight peak at 350 before the panel became flat. This is where active analog crossovers really comes in handy. A quick turn of the dial while watching the RTA would simplify finding the exact crossover point. Then a sweep to see how well the sound transferred between drivers could verify this.
The end result was far beyond my expectations! I figured I'd have a better stage and more defined midrange, I did not expect to cure the sound shift that occurs while standing up and moving around. That's no longer an issue, when sitting in the golden spot.
I don't know any specifics about JansZen panels so I can't answer that question.
I believe that diaphragm coatings are the weak point in most ESLs so I will speculate that JansZen's longevity must result from using a very robust diaphragm coating.
For example; Acoustat used an essentially bullet proof brushed-on laquer/carbon black diaphragm coating, and many old Acoustats still playing fine 35 years on. The carbon black coating is relatively heavy and completely opaque so would not be suitable for a see-thru ML panel but perhaps JansZen used a carbon black coating... just speculating.
Last edited:
So you think Martin Logan values the transparency of their panels enough that they use a less robust coating? I could care less if mine are see through. I'd much rather have the dark coating.
Yes.
I think aesthetics is the priority for marketing purposes but their see-thru coating likely has more optimal (higher) resistance than a carbon black coating. Also, higher resistance limits charge migration across the surface, allowing the panel to operate closer to the ideal contant charge mode.
Arcing wasn't a problem on the Acoustats becuase their wire stators were so well insulated.
I'm just speculating here... but for purely electrical reasons, a carbon black coating may not be a the best choice in a ML panel, even if one had no objection with its opaque black color.
Last edited:
But when I compare them to a ML panel that's half their age and I also have 4 different versions of JansZen panels to compare with, to say that I don't have anything to compare to I have to disagree. See this is why I had to ask this questionThe big factor I have not seen one person mention here is this. If you find a pair of lets say Ascent-i speakers that are 20 years old and you’re thinking of buying them so you go listen to them and you say wow those sound great why would the panels need to be replaced? Well here is my point, when you listen to the 20 year old speakers you DO NOT HAVE A PAIR OF NEW PANELS to compare the 20 year old panels to, and from everything used I’ve listened to there is a discernible difference the vast majority of the time, my 20 some year old Odysseys were definitely that way. Yep they sounded good with the original
panels, but with the new panels they sound friking superb! A large noticeable difference. I know nothing about the theories as to why they wear out but they do from everything I’ve heard. My Odysseys had one speaker with just slightly less volume output than the other, new panels installed perfect balance. I totally respect Jazzman’s comments he’s put countless hours in with building, researching and understanding electrostatic speakers. This is just my 2 cents worth.Take it for what you will.
Since I saw no noticable sound loss between 4 sets of JansZen designed panels with mfg dates of 1959, 1967 and 1971 (especially in the highs) I was confused how the MLs would lose so much of their high frequency output?
By the way people who wash them end up with delamination on the sides, I have only to think that the panels sides are coming loose while staying in place. Which is more of a manufacturing defect from adhesive breakdown than anything that pollution could cause.
Of course this is only based on what I've read. The last thing I'd ever do is cause the panels to lose tension. So washing them to me is a huge risk.
I have found out through my own research in positioning panels in various formations, that placing them in a curved configuration causes the staging and imaging to fall apart. The most curvature that can be used without losing the magic of an ESL is to aim 1 out of 4 panels in a horizontal plane and have the outer panel aimed slightly towards the wall. This keeps 75% of rear waves on a straight path to reflect off the rear wall. The outer panel is reflecting off the side and will fill in the empty or less populated area in between the right and left speakers while blending partially back with the front waves.
The 25% panel dispersion is actually about 10% reflected sound so the ability to enlarge the sweet spot can only be overcome by a wider panel in a horizontal plane. Crossing the rear waves is a huge no no. I tested this extensively.
I also tested the blending of transitional effects at the crossover point and found the high or steep slopes used by most makers to be the problem, not the fix. The more the 2 lap over each other the less likely it is to hear the transition at the crossover point. What this needs is a woofer that can closely match the timber and level of output needed at the crossover point to be successful.
Working with the drivers natural roll off or even better dips in output can used very successfully to become a crossover/lap point where a 6db slope is going to be best. I was able to find characteristics like this by testing woofers and then testing 2 woofers of different size. The difference in size offers a few changes to output. There can be peaks and valleys show up and used to the advantage of the mixture with the panels natural roll off or dip it may have at some point.
Of course I did this on a 1 off build, I had no care of reproducing this design. Once I lost the perfect pair I made I was once again looking at finding something that was close to what I knew was possible but found the fundamental basics (I discovered) not being part of designs.
I was able to do a fix on the Aerius and have posted my work here for others to see. The removal of the crossover and going active was a big step forward but it couldn't help the poor imaging and staging. What helped that was the addition of another woofer mounted on the back cover.
The reinforced rear waves have solidified the imaging to a point where it sounds like I have new speakers. It's not the answer but it's the only fix I could come up with while using a curved panel design which is flawed from the start.
I originally had the idea to use my 28 JansZen panels along with 2 Mirage BPS150i subwoofers that use 2x 8" woofers. I bypassed the amp on the one I owned and ran an externally controlled amp with the speakers aiming, one forward and the other towards the rear wall.
The sound was amazing. I could not find another matching sub and used other subs and after a few months of testing found the way to go forward. The bass was super clean and powerful. The Mirage way would have been much easier and had a very nice look.
Right now I'm using those same mirage drivers in the back of the Aerius. I have changed out the front firing drivers as well with another brand that's an 8Ω driver. I do have 4 matching drivers I will be testing later today. The increase in output was a low frequency of 43Hz at reference with pure tones and extension to the 20s with white noise or music. Since I'm running amps that can do 1Ω without a problem a 4 ohm load is not even making it hot. I have considered running 4 BGW A150s all in mono (one for each driver) but I need to make some new XLR cables to do it. I know I have the output voltage from my crossover but I haven't had time to start making the cables till now. I think the 4 amps will be able to control the speakers better than 1 amp. The BGWs have a reasonably high damping factor and my wires will only be a few inches in length since I can place the amps right behind the cabinets. That will keep the damping high and allow 150w per driver.
Current output is actually more than the BGWs but the damping will be less, plus I don't push the speakers passed what I consider to be reference sound, and why would I? I tried the amps running in stereo and found them to be under powered for use with the Aerius speakers. They are rated at 75w x2 at 4 ohms and are rated to power ELSs without any issues.
I ran 4 sets of ESLs in parallel running on QSC 3500 series 3 amps without any problems.
BGW claims they can run any load and are rated for ESLs, they are air cooled like the QSCs, I have to get the bias dialed in on one of the QSC amps, it's a complex process that requires more pieces of equipment than I possess, I haven't researched for anyone local who can do it. I tried it by matching voltages at the adjustment pots but it still gets hot. I've tested everything else and found nothing that's out of spec. The heat is spread out so no single part is causing it.
This was my last sweep plot.
Not too bad. Compared to the first one.
