Resistors may not be futile, Star Trek Borg quote (somewhat)
I am a believer in cap replacement, but not just to replace tired ones. In some instances I think upgraded caps can make a well designed unit sound better. I have done two Luxman units, TP117 and R117, complete using audio grade and both turned out sounding much better than the originals. This pair takes the place of some more highly regarded, more expensive, separate units that I own. I own several of each of the Luxman units in the original states, and so I am not relying on memory when I say they sound better. Currently I am repairing one of my R117s that had blown transistors on the L Amp board. I am doing a recap of the board and at the same time I am scratching a long time itch and replacing all of the the amp related resistors with Vishay-Dale low noise metal film units. I am doing the input signal path where needed as well. If I can tell a difference between the recapped, original and the recapped with resistor replacement I will let you know. Oh and yes the R Amp board and tone control boards will need to be done to make the test legit.
I am a believer in cap replacement, but not just to replace tired ones. In some instances I think upgraded caps can make a well designed unit sound better. I have done two Luxman units, TP117 and R117, complete using audio grade and both turned out sounding much better than the originals. This pair takes the place of some more highly regarded, more expensive, separate units that I own. I own several of each of the Luxman units in the original states, and so I am not relying on memory when I say they sound better. Currently I am repairing one of my R117s that had blown transistors on the L Amp board. I am doing a recap of the board and at the same time I am scratching a long time itch and replacing all of the the amp related resistors with Vishay-Dale low noise metal film units. I am doing the input signal path where needed as well. If I can tell a difference between the recapped, original and the recapped with resistor replacement I will let you know. Oh and yes the R Amp board and tone control boards will need to be done to make the test legit.
Actually you are "tuning" the amp... did you simulate it in eg. LTSpice first?
It seems to me, that using a simulation-tool you can achieve an optimal result.
Here is an ESR tester that is currently available. It is said to be quite good, but I don't have one myself, so I can't vouch for it (I would like one, but finances do not permit it at present).
More info at
Peak Electronic Design Limited - Atlas ESR - Equivalent Series Resistance Meter and Capacitor Analyser - Model ESR60
R 1,403.48 from RS components
Do you know of a cheaper ESR meter for sale in South Africa?
EDIT:
R 1,403.48 is for ESR60 Meter
They also cell ESR70 Meter for R 1,298.27
So the better model is actually cheaper
Regards
Last edited:
Look on aliexpress for ESR meter.. a module* without case for approx. USD 25,- including shipping...R 1,403.48 from RS components
Do you know of a cheaper ESR meter for sale in South Africa?
EDIT:
R 1,403.48 is for ESR60 Meter
They also cell ESR70 Meter for R 1,298.27
So the better model is actually cheaper
Regards
*) about the same functionality and accuracy, but tests also inductance, diode, FET, pnp and npn
Last edited:
Many of the cheap ESR meters are not suited for serious work especially on larger caps (like power supply smoothing caps) where ESR is low. This is for a very simple reason, they do not have compensation of test lead resistance. Anything below 0.2 ohms or so should have separate current supply and voltage sensing leads or most of the ESR you will measure is the contact resitance plus test wire resistance. Good smothing caps will often have ESR in 10s of miliohms. Compensating for lead resistance by subtracting the resistance of shorted test leads will still lead do huge errors (like 50%) becaue of the different contact made between leads vs between lead and capacitor terminal.
Since we are in the solid state forum, I will not go into some stuff which needs replacing on vintage tube equipment such as carbon comp or even some types of carbon film resistors, not to mention paper (not paper in oil) caps.
1) Electrolytic capacitors - these have the lowest life expectancy of all parts (not counting parts which are unduly strained and work at the edge or even beyond maximum spec). There are advantages in using different types of electrolytic compared to what was originally used, as cap technology has progressed in the meantime. More on that below.
There are also known cases of particular non-electrolytic capacitor constructions that are known to fail. An example of this are so called 'black flag' silver mice caps used in many high end vintage equipment in the early 80s, which have an internal contact corrosion problem and can abruptly turn into 'nothing'. Since these were often used for compensation, amplifiers would suddenly and auto-destructively turn into oscillators.
2) contact gear of various kinds such as connectors and relays - often these can corrode or in case of relays have been exposed to destructive switching cycles (for instance speaker protection relays in situations where the output stage has been blown and full DC went momentarily to the load - arcing and material transfer can occur because of DC current switching).
3) Problematic semiconductors. Semiconductor technology has advanced in terms of reliability, and this does not necessarily (and in fact normally does not have to) do with reliability of the actual semiconductor, rather it's case. Early plastic enclosed diodes and transistors often had problems where terminal corrosion would lead to the innards getting exposed to the atmosphere. Some parts are specially prone to failure, such as small (often multiple junction) diodes that look like a small ball with two thin wires. These tend to go open with no warning. In some cases semiconductors were used right at the limit, resulting in various constant or intermittent problems such as leakage currents, vastly increased noise, or abrupt failures with no apparent explanation.
4) Resistors. Apart from carbon composition, which is fairly uncommon in solid state vintage, there are types that are known to be problematic, such as fusible resistors. These combine aspects of a PTC in order to act as a fuse when a current over a certain limit is passed through them. The main idea is to prevent burning, i.e. to be flame retardant or resistant, a problem with older carbon film based resistors, as carbon will readily burn. It turned out that the fusible resistors tended to increase resistance quite drastically outside of their tolerance rating or even go completely open even within operating conditions that would be considered well inside spec, given enough time. These days they are replaced with metal film or metal oxide resistors which are normally flame retardant or resistant (depending on actual construction).
5) Trimmer potentiometers, which usually develop mechanical contact problems due to corrosion, and less often, mechanical problems (also usually due to corrosion). In some cases they were used in such a way that intermittent contact on the wiper can lead to destructive failure of parts of the circuit, most commonly this is found on bias settin trimmers.
Upgrading - this is always a difficult question as in some cases it cannot be helped since original components are not available. It is reasonable to replace the original with something at least equal in spec if not better. However, it is not an easy matter if one wants to preserve the original aspect of the equipment, including it's sound signature. That being said, we are talking about equipment which is upwards of 20 years old. It is safe to say that NONE of it sounds like when it was new, so comparison of restored state and non-restored state is a moot point. The only safe thing to say is that if it doesn't sound better once restored (given a proper break-in period), then something is wrong. Also, experienced restorers have a good feel about the sonic signature, or 'house sound; if you will, of components of certain brands, and sufficient knowledge of the circuitry to know what parts can be used as replacements when original spec parts are not available, and where. This is NOT easy, far from going through the parts list and ordering all new caps from the best brand or boutique manufacturer of one's choice. There are a lot of such things out there, where people will put expensive boutique electrolytic caps into a DC protection circuit or a auxiliary supply that drives relays and front panel lamps - and yet miss real opportunity to upgrade performance without compromising the original philosophy of the circuit.
There are two classes of upgrades, and one should be made and should be considered added value to a vintage piece - these upgrades have to do with reliability, i.e. replacing parts that are known to be problematic, prone to failures of degradation of specs, such as some mentioned above (fusible resistors, certain kinds of relays and sockets, semiconductors, black capacitors). Restoring a vintage piece with NOS parts that are known to fail is really pointless, regardless of the fact that unmodified equipment is usually considered more valuable. These kind of mods are often what the manufacturers themselves would publish in service bulletins and would do themselves if they had known of the problems in advance.
The other kind of upgrade can be using upgraded parts or even circuit modifications, to upgrade the basic specification of the equipment - or to attempt to (often unsuccessfully!). Someone mentioned the use of a simulator to investigate such possibility and this is indeed a good tool to use. However, it is often even better to look into other product of the same manufacturer and line or family, to find clues about the design philosophy of a piece of equipment. For instance, if you see a coupling electrolytic cap in parallel with a film cap, or a bipolar electrolytic used, it is safe to say that the designers would have used a film cap if they could fit one or have it at a target price. These days getting small film caps up to 4.7uF (and sometimes 10uF) is far less of a problem and using one instead of an electrolytic type is almost always a safe bet. In the same way, wherever there is no or very little DC (less than 0.1V) across an electrolytic, using a bipolar or audio electrolytic (which are designed to work well under these conditions) is a good idea. In many cases power semiconductors were used close to their limits, often the very same ones were part of a family of semis that continued to be developed into newer and more robust or higher spec reincarnations which are natural replacements for the old and frequently obsolete parts.
In my case, i am very much opposed to radical changes and mods. It does take knowledge and experience to figure out what the original designers were up to and what they wanted to achieve, but it is also a great learning experience when you do. IMHO it is important to retain the design philosophy and always ask what the original designers would do given better parts before doing any upgrade. That in my mind is the only way to keep vintage equipment vintage, and not an attempt of a replica or something altogether different in a vintage skin. I do have one exception to this rule - and that is saving a badly damaged or mangled piece of rare vintage equipment, but even in that case I will try to re-manufacture parts to get it to original state, or at least keep the 'core' of the device as original as possible - for instance I had a case of power amp that had the entire front panel destroyed, including controls and vu-meters. These cannot be replaced except from a donor amp, but considering the rarity, this is unlikely. However, the actual amp works and is original - I did not try to do a replica of the front panel as that part does not do anything for the sound, rather repackaged it and operated it as such.
1) Electrolytic capacitors - these have the lowest life expectancy of all parts (not counting parts which are unduly strained and work at the edge or even beyond maximum spec). There are advantages in using different types of electrolytic compared to what was originally used, as cap technology has progressed in the meantime. More on that below.
There are also known cases of particular non-electrolytic capacitor constructions that are known to fail. An example of this are so called 'black flag' silver mice caps used in many high end vintage equipment in the early 80s, which have an internal contact corrosion problem and can abruptly turn into 'nothing'. Since these were often used for compensation, amplifiers would suddenly and auto-destructively turn into oscillators.
2) contact gear of various kinds such as connectors and relays - often these can corrode or in case of relays have been exposed to destructive switching cycles (for instance speaker protection relays in situations where the output stage has been blown and full DC went momentarily to the load - arcing and material transfer can occur because of DC current switching).
3) Problematic semiconductors. Semiconductor technology has advanced in terms of reliability, and this does not necessarily (and in fact normally does not have to) do with reliability of the actual semiconductor, rather it's case. Early plastic enclosed diodes and transistors often had problems where terminal corrosion would lead to the innards getting exposed to the atmosphere. Some parts are specially prone to failure, such as small (often multiple junction) diodes that look like a small ball with two thin wires. These tend to go open with no warning. In some cases semiconductors were used right at the limit, resulting in various constant or intermittent problems such as leakage currents, vastly increased noise, or abrupt failures with no apparent explanation.
4) Resistors. Apart from carbon composition, which is fairly uncommon in solid state vintage, there are types that are known to be problematic, such as fusible resistors. These combine aspects of a PTC in order to act as a fuse when a current over a certain limit is passed through them. The main idea is to prevent burning, i.e. to be flame retardant or resistant, a problem with older carbon film based resistors, as carbon will readily burn. It turned out that the fusible resistors tended to increase resistance quite drastically outside of their tolerance rating or even go completely open even within operating conditions that would be considered well inside spec, given enough time. These days they are replaced with metal film or metal oxide resistors which are normally flame retardant or resistant (depending on actual construction).
5) Trimmer potentiometers, which usually develop mechanical contact problems due to corrosion, and less often, mechanical problems (also usually due to corrosion). In some cases they were used in such a way that intermittent contact on the wiper can lead to destructive failure of parts of the circuit, most commonly this is found on bias settin trimmers.
Upgrading - this is always a difficult question as in some cases it cannot be helped since original components are not available. It is reasonable to replace the original with something at least equal in spec if not better. However, it is not an easy matter if one wants to preserve the original aspect of the equipment, including it's sound signature. That being said, we are talking about equipment which is upwards of 20 years old. It is safe to say that NONE of it sounds like when it was new, so comparison of restored state and non-restored state is a moot point. The only safe thing to say is that if it doesn't sound better once restored (given a proper break-in period), then something is wrong. Also, experienced restorers have a good feel about the sonic signature, or 'house sound; if you will, of components of certain brands, and sufficient knowledge of the circuitry to know what parts can be used as replacements when original spec parts are not available, and where. This is NOT easy, far from going through the parts list and ordering all new caps from the best brand or boutique manufacturer of one's choice. There are a lot of such things out there, where people will put expensive boutique electrolytic caps into a DC protection circuit or a auxiliary supply that drives relays and front panel lamps - and yet miss real opportunity to upgrade performance without compromising the original philosophy of the circuit.
There are two classes of upgrades, and one should be made and should be considered added value to a vintage piece - these upgrades have to do with reliability, i.e. replacing parts that are known to be problematic, prone to failures of degradation of specs, such as some mentioned above (fusible resistors, certain kinds of relays and sockets, semiconductors, black capacitors). Restoring a vintage piece with NOS parts that are known to fail is really pointless, regardless of the fact that unmodified equipment is usually considered more valuable. These kind of mods are often what the manufacturers themselves would publish in service bulletins and would do themselves if they had known of the problems in advance.
The other kind of upgrade can be using upgraded parts or even circuit modifications, to upgrade the basic specification of the equipment - or to attempt to (often unsuccessfully!). Someone mentioned the use of a simulator to investigate such possibility and this is indeed a good tool to use. However, it is often even better to look into other product of the same manufacturer and line or family, to find clues about the design philosophy of a piece of equipment. For instance, if you see a coupling electrolytic cap in parallel with a film cap, or a bipolar electrolytic used, it is safe to say that the designers would have used a film cap if they could fit one or have it at a target price. These days getting small film caps up to 4.7uF (and sometimes 10uF) is far less of a problem and using one instead of an electrolytic type is almost always a safe bet. In the same way, wherever there is no or very little DC (less than 0.1V) across an electrolytic, using a bipolar or audio electrolytic (which are designed to work well under these conditions) is a good idea. In many cases power semiconductors were used close to their limits, often the very same ones were part of a family of semis that continued to be developed into newer and more robust or higher spec reincarnations which are natural replacements for the old and frequently obsolete parts.
In my case, i am very much opposed to radical changes and mods. It does take knowledge and experience to figure out what the original designers were up to and what they wanted to achieve, but it is also a great learning experience when you do. IMHO it is important to retain the design philosophy and always ask what the original designers would do given better parts before doing any upgrade. That in my mind is the only way to keep vintage equipment vintage, and not an attempt of a replica or something altogether different in a vintage skin. I do have one exception to this rule - and that is saving a badly damaged or mangled piece of rare vintage equipment, but even in that case I will try to re-manufacture parts to get it to original state, or at least keep the 'core' of the device as original as possible - for instance I had a case of power amp that had the entire front panel destroyed, including controls and vu-meters. These cannot be replaced except from a donor amp, but considering the rarity, this is unlikely. However, the actual amp works and is original - I did not try to do a replica of the front panel as that part does not do anything for the sound, rather repackaged it and operated it as such.
Absolutely true! But that's the problem you get when you don't have the availability (financial means) of lab-grade equipment...
BTW: Thanks for your post on upgrading.
Forgot to mention that these testers are still quite useful, if they can measure capacitance to a fairly accurate degree (say 5% or less error).
It should be noted that measuring power supply filters for pass/fail is quite different than the rest of the capacitors in a vintage unit (especially amplifier). Main filters will rarely just have increased ESR, rather the capacitance will decrease - quite often without a significant rise in ESR. I will not go into the mechanisms of that particular failure (or rather, simply being worn out) but it's a good idea to simply look at the capacitance tolerance figures given on the capacitor - usually up to -10% on capacitance is tolerated right out of the factory. If the capacitance is within that limit, in 99% of all cases it can be assumed that the cap is still OK and able to perform it's duty as a smoothing cap.
It is not easy to make this assumption for other capacitors in the circuits. Local DC smoothing or decoupling caps, also in most cases DC block or even coupling caps with permanent DC (at least a volt or so) operate under different conditions, those being the absence of ripple current. These can indeed retain capacitance within tolerance limit but develop higher ESR (especially if they are close to components that heat them up). Modern caps are better at this right out of the factory and will often provide better service without going exocit on types to use.
In many cases there are electrolytic caps used in places where there is very little or no DC across them, even potentially small reverse DC. These can develop leakage which can impact circuit performance quite a bit. I always replace these with foil caps wherever I can, unless the capacitance is too high, in which case a good bipolar cap will do the trick. FOr very large capacitances where even bipolars are impractical or simply not available, a good audio cap will do - and in many cases a properly polarised general purpose cap (Panasonic FC comes to mind) - even a few mV in the properly polarized direction is better than nothing or reverse polarization (although care should be taken to check that these mV are there due to proper circuit operation and not some failure of another parts such as a leaky transistor, as manufacturers will generally take care of these things and properly polarize caps to account for it).
I am about to purchase this one.
Der EE de 5000 High Accuracy Handheld LCR Meter with TL 21 TL 22 New Japan | eBay
Der EE de 5000 High Accuracy Handheld LCR Meter with TL 21 TL 22 New Japan | eBay
Statistics say NO ...these hardly ever fail
Though in specific point in class AB amplifiers there is capacitors that will perform better than a smiple ceramic .
Notice that it is not enough to replace with something else that has the same capacitance Voltage is critical and the "make" of the capacitor is also critical
A riaa preamp will be far better with more accurate capacitors MKt or mica
Feedback circuits and filter circuits will love styroflex or mica capacitors Miller capacitors also alike
Choose with wisdom
Nah, the ceramic caps in my 1964-68 Hammond organs are fine.
I wouldn't say that the ones in some bargain TV's of the same age were fine, but those things have been scrapped. the format is obsolete. Certain vendors bought the cheapest trash they could find, houses like Scott, Dynaco, MacIntosh, Hammond Clock bought the good stuff. Which is why I bought the equipment you see in my tag line. Survivors that went 20 years (to my purchase in 1970) don't have trouble going 30 more years, except for the **** rubber sealed e-caps. Put 72 e-caps in one of the H182's, 115 to go in the harp circuit. You can hear the results on inbojat.tumblr.com the yellow bird track.
To Enzo's point, ceramic bypass caps in power supplies , miller compensation, and anti-oscillation around amp stages have to be ceramic or silver mica. Wound caps like any plastic film wound cap has too much self inductance to act at very high frequencies. Stacked plastic may begin to be useful, haven't seen any discussion of them.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.