• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Vintage Speaker Baseline Measurements Before Re-cap/Modification/Repair?

CanadIanGeek

New Member
Short Version: I want to know what basic measurements I should take (before and after making changes) to get a decent picture of how a given speaker is performing beyond what my less educated ears tell me.

Long Version: I have 4 sets of vintage speakers from the 1970s that really should get a re-cap (and some need box repairs) to ensure they play long into the future. Over the years I've bought all the equipment and parts I need to do the job, but every time I take a step closer I get cold feet and walk away. This is mostly because of the complexity of a system where a single change can have unforeseen consequences. Being that the re-cap work is pretty reversible I want to be able to check my work by taking measurements before and after. I'm pretty sure I have enough equipment to do the measurements I need, but I just need to know what measurements I should take so that I can then educate myself on how. I don't have access to an anechoic chamber, and can't easily do the measurements outside.

Equipment:
  • DATS V3: I bought this because it gave me a tool for measuring components(capacitance, inductance) check drivers, and bonus it can be used to measure speakers and check for box issues etc.
  • MiniDSP UMIK-1: Calibrated measuring mic. I caved and bought this when I realized I could use it for my various speaker repairs and modifications, but could also use it to calibrate my Pioneer surround receiver.
  • Multi-Meter: Might as well list this here since it does measure and I do have it. It's a basic digital unit, nothing fancy.
Questions:
  1. What software should I be using?
  2. What measurements should I be taking?
  3. What checks should I be doing with the individual drivers?
The Speakers(with what work I intend to do):
For those who are about to scream about me ruining a bunch of precious old artifacts of a bygone age, I will cut you off at the pass. None of my speakers are models that are revered for their audio perfection. I love them and intend to stay true to what they are, but they are in need of some love. The better the condition they are in the less I intend to do. I'm not striving for perfection or trying to flatten their response curve nor do I expect night and day differences. I'm just trying to stabilize things and sand any rough edges. If no one notices a difference that is a success, if I notice a tiny improvement that is a huge bonus! If things get worse, I want to know right away, so that I can roll back the changes or go a different direction.
  1. Two pairs of Celestion Ditton 25: One of these is a later model they were my first Hi-Fi equipment(high sentimental value). They need some major box repairs(I lived in a bus for 6 years), but they still sound fabulous. The other set is in VERY good condition, but the capacitors look sketchy as hell! These have a reputation for having very sensitive, much praised and sought after, tweeters that blow if the the crossover sends it too low a frequency. The two sets are from different build generations, they have the same drivers but the later model(with the damaged boxes) have MKP caps where the other ones have hard to identify caps that are covered in a white precipitate. So the fact that Celestion upgraded feels like it gives me full license to do the same :P !! These are going to be the first ones that I work on because I don't want them in service until those caps get changed. Right now they are a waste of space in my living room so no procrastination allowed!
  2. Leak 2060: These are on my desk(it's a big desk)! These get used every work day. Love them, but the whole crossover is electrolytic caps that are VERY old. They also have almost no stuffing in them, so I might experiment with changing that, but I want to change as little as possible.
  3. Akai SW175: These are in my big room(kitchen/dining room). These have the cheapest looking crossovers and they have only a bit of fiberglass less than a cm thick stapled to the sides of the box. The box also has a hole through(from the factory) where there is a 1/4" headphone jack(I guess they used it as an alternative quick connect cable). These also have a reputation for having too small a box for the big 15" woofer, so it could probably be improved with some proper stuffing work and maybe seal the hole! They sound shockingly good, if a bit weak on the bass side.
 
Register to hide this ad
REW is free software that runs on a PC. You can use it with the Umik to run frequency response sweeps. Frequency response is the main measurement you’ll need to see if installing new caps makes a difference. But your ears will probably tell you the same thing.
 
Last edited:
Having Fast Fourier Transform (FFT) to analyze is great since speaker crossovers are reactive (both capacitive and inductive) and you may be able to see differences in not only the fundamental, but harmonic structure of the speaker system.
 
I see lots of post on the forum which imply that taking measurements is easy. I disagree. It's easy to get misleading results and inconsistent readings, particularly at the LF end of the spectrum. I use ARTA, and there's a learning curve involved.
I'd change the ancient capacitors and see what you think. Let your ears guide you. Looking at graphs could end up driving you mad.
 
My own opinion, and this is indeed just my own opinion, is not to overthink it. Box repairs and/or driver repairs like refoaming are a given. If a speaker is sealed and the box is leaking, it won't be working as designed. Likewise, if the joints are rattling, this will add resonances. If the drivers need refoaming, and you are using the foams from recommended suppliers (Tom at MyAudioAddiction, Rick Cobb AKA looneytune2001, etc.) then that is the best you can do. If you do not feel up to the job, then use a speaker repair professional (others can advise here).

So that brings us to caps. Since you have two speakers, replace the caps in one, and set it up side by side with the other one, and play music in mono through both of them. If the recapped one sounds better, great! You can then replace the caps in the other. If it sounds worse, put the old caps back in. If it sounds the same, consider how much life the old caps have in them. Would you rather put in the new ones, and figure they will last for another 20 years? Or leave the old ones in, and wonder whether they will last the next two years, or 20 years.

You can also play around with adding or removing some of the stuffing inside the boxes. But in my opinion, the best measuring devices are on either side of your head. You are the one living with the speakers, so you have to like the sound. Supposing, you add stuffing, and the measurements tell you the bass response is flatter. What if your ears are hearing that bass hump that used to be there, and liking it?
 
Or you could just do one individual speaker at a time and compare them side by side with your own ears (the most important measuring device).
 
Thanks for the feedback!

I'm definitely over thinking things and procrastinating :)
The first recap job is the set of first generation Celestion Ditton 25. This is also the set that has the most info from people who have worked on them. The general advice is to replace all but the capacitor across the subwoofer with MKP or similar, but to replace that one with a new Alcap low loss electrolytics (not sure why). Here in Canada those are a bit complicated to order, and I was able to get the correct value MKP with my order of all my other caps. I bought them, but now I'm overthinking the whole thing and feel like having measurements will give me a better picture of whether I have done something weird to the crossover.
I agree that if it sounds good then it's good, but a measurement can quickly tell me if something is bad that may effect a limited range of frequencies so may be hard to pick out from the mix(with my less trained ears) or just not be used by certain recordings.
A basic frequency sweep may be enough for my purposes.
 
The first thing to do is to document the existing conditions. This includes things like the volume levels on the equipment and any Lpads, exact placement and location on the floor, changes in room furnishings, etc. Then do your measurements such that you can ensure that there are no differences other than the changes that you made to the crossovers when you put things back to re-measure.
 
REW is free software that runs on a PC. You can use it with the Umik to run frequency response sweeps. Frequency response is the main measurement you’ll need to see if installing new caps makes a difference. But your ears will probably tell you the same thing.

I also suggest giving REW a try. Your UMIK-1 will pair with the software perfectly. I did go to YouTube for a tutorial to get me started.
So far I've used it to do sweeps and make some corrections to the room response.
 
Back
Top Bottom