I collect old radios and lots of them even the cheap models have chokes. I wonder why, were large caps so expensive back than that choke was cheaper to use? I don't think it was for hum suppression because I never use chokes in amps I build and they're totally quiet.
Before electrolytic caps, chokes were probably simpler and smaller that capacitors.
Possibly more reliable than paper in oil or similar things.
Possibly more reliable than paper in oil or similar things.
Electrolytics used to have small values - maybe 8uf reservoir and 16uF smoother. Much more than this became too big and too expensive. A choke makes up for this. Also, small reservoir caps are a better fit to vacuum rectifiers. It is simply a matter of designing the best compromise given what technology is available at reasonable cost.
Nowadays we have big electrolytics and SS rectifiers (which can handle the large peak pulses). Therefore we have less use for chokes, which is fortunate as they are expensive. Remember, radio sets were expensive in the old days - a house might only have one or two of them and some of these would be cheap ones which hummed.
Nowadays we have big electrolytics and SS rectifiers (which can handle the large peak pulses). Therefore we have less use for chokes, which is fortunate as they are expensive. Remember, radio sets were expensive in the old days - a house might only have one or two of them and some of these would be cheap ones which hummed.
What PSRR your gear has?
How big your psu ripple is? ~ (m)V
Being hum free, and low psu ripple is not the same. Audio signal gets modulated by ripple, and this gets audible. (addition-substract frequencies)
I get 1Vrms psu noise on 1st stage, 2nd stage has 80mVrms. //400VDC
1/2ft of wire hanging on probe shows about 140mVrms😀
Attachments
Hi,
In the old days the valve rectifier tubes could not cope with
the turn on current of large filter caps, that is what limited
capacitor values to being quire modest, hence CLC filters.
More to do with the cost of good rectifier tubes than caps.
rgds, sreten.
In the old days the valve rectifier tubes could not cope with
the turn on current of large filter caps, that is what limited
capacitor values to being quire modest, hence CLC filters.
More to do with the cost of good rectifier tubes than caps.
rgds, sreten.
Last edited:
No. If that were the case then most sets would use the maximum cap value allowed by the rectifier. They did not.sreten said:
For example, the K-B BR20 (5Z4 rectifier - max cap 40uF) uses 16uF. The McMichael FM55 (EZ80 rectifier - max cap 50uF) uses 32uF. The Pye P45 (EZ40 rectifier - max cap 50uF) uses 24uF. That is just some radios for which I happen to have the service sheets because I own or owned them. As it happens, none of these used a choke but either a CRC filter or the OPT primary used as a hum canceller. Thus they could have benefited from bigger caps and the rectifiers would have coped but they didn't do this.
It was cap size and cost which limited things.
To a point. Enough C behind enough R could still get you there. High capacity/Voltage electros used to be large, expensive and unreliable.
Edit: @ sreten
Edit: @ sreten
crc filter use for low current circuits like radio .but for more current how use crc?in olde radio used sensitivity spekaers .and hum can hearing . for reduce and reject hum in speaker crc help but kind of wiring is most important !!!!in vintage guitar amp clc and crc can not solve hum .but used kind of Assembly and wiring to kill hum
I don't know about you guys, but i use a LCLC sometimes cLCLC filter for nearly all my single ended amplifiers. The caps i use are mostly film caps that are less than 100uf. If you check out the old radio books like the "Radio Amateurs Handbook" in the power supply section, the LCLC filter is the best filter. Sure chokes are more expensive but they are better sounding than a dropping resistor and allow the builder to use smaller value caps for the same filtering results. This makes your power supply have lower impedance, than with large filter caps which can result in a better sounding amp.
The last amp i made was a SE 6550 with a cLCLC filter or .5uf vitaminQ x 4H X 20uf pio x 8 Henrys x 70uf film cap . It is dead quiet in an Altec 19 speaker of 105db sensitivity.
I won't comment on the "better sounding" claims, but it is indisputable that a modern electrolytic cap of 1000uF offers much lower impedance than a 70uF film cap at 20Hz. Also, by using big 'lytics in R/C filters with filtering performance equal to L/C filters with smaller caps, one can achieve equal or better regulation because the resistor values don't have to be any greater than choke DCR.
Resistors ; sure you have some cooling on them too.
http://www.mouser.com/ds/2/177/EDB194B-69609.pdf my chokes are 108DCR, 4H = XL @100Hz 2513R
Amp draw 80mA @ 108R= cold to touch. And only <10VDC drop. 2513R resistor would drop >200V,16W heat
You can throw thousands µF at a problem (ripple will be decreased slowly), but do not forget about safety, where you must discharge them in short time. Such supply is harder on diodes and trafo. Bigger ripple current - bigger hum risk; capacitors have max ripple current limit too
You can use fet filter (with more voltage drop to achieve filtration), heatsink probably a must. Fets does explode/short when overloaded
http://www.mouser.com/ds/2/177/EDB194B-69609.pdf my chokes are 108DCR, 4H = XL @100Hz 2513R
Amp draw 80mA @ 108R= cold to touch. And only <10VDC drop. 2513R resistor would drop >200V,16W heat
You can throw thousands µF at a problem (ripple will be decreased slowly), but do not forget about safety, where you must discharge them in short time. Such supply is harder on diodes and trafo. Bigger ripple current - bigger hum risk; capacitors have max ripple current limit too
You can use fet filter (with more voltage drop to achieve filtration), heatsink probably a must. Fets does explode/short when overloaded
I subscribe to this at lower voltages but good luck trying to find a 1mF cap rated for 500vdc. Chokes last much longer than lytics' too.
Low value caps mean a high impedance supply. A choke input supply can be low impedance, but only if the choke has the right value and rather low resistance. Many audio people completely spoil this by adding a small input cap (call it cLC filter?) which gives a higher impedance supply than either choke input or cap input; the tradeoff for this awful performance is that you can adjust the DC voltage by varying the cap - which in most cases is quite unnecessary. So you lose something you need (low impedance) in exchange for something you don't need (exact DC voltage).DAK808 said:
Here's an example to illustrate my point: An L/C section with 4H 108R choke and 100uF cap attenuates 100Hz ripple by a factor of 15.92 / 2513 = .00633. Two R/C sections of 47R and 1000uF attenuates 100Hz ripple by (1.592 / 47)^2 = .00115 and has lower DCR.
SMPS maybe has opened up a new door for high voltage high value caps?
I haven't looked on Mouser for high voltage caps in a while, they are big and somewhat expensive.
http://www.mouser.com/ProductDetail...=sGAEpiMZZMvwFf0viD3Y3bzn7LR4wbx4PIu/6YS/NXg=
I haven't looked on Mouser for high voltage caps in a while, they are big and somewhat expensive.
http://www.mouser.com/ProductDetail...=sGAEpiMZZMvwFf0viD3Y3bzn7LR4wbx4PIu/6YS/NXg=
Of course in the old days at least one radio in the house used an accumulator which had to be recharged. My mother told me how she used to have to take it to the shop to get it recharged.
Look at this ridiculous cap.
http://www.mouser.com/ProductDetail/Panasonic/EZT-VKCTYP1HA/?qs=3ADrEY1raEV2MOepIse%2fxw%3d%3d
I can see some audiophile putting this into their 300b amp. It only costs $285.
I suspect people put film caps at the decouple position feeding the OPT in SE amps, well that's where I put them. @ 20Hz the film cap I linked above had and ESR of .4 and the lytic I linked in post #15 had and ESR of .6
http://www.mouser.com/ProductDetail/Panasonic/EZT-VKCTYP1HA/?qs=3ADrEY1raEV2MOepIse%2fxw%3d%3d
I can see some audiophile putting this into their 300b amp. It only costs $285.
I suspect people put film caps at the decouple position feeding the OPT in SE amps, well that's where I put them. @ 20Hz the film cap I linked above had and ESR of .4 and the lytic I linked in post #15 had and ESR of .6
Last edited:
True for the input side from the rectifier of the filter . Then look at the impedance of the filter at higher freq.s in the audio range as it relates to the circuit needs for power . the resistor is some what flat while the choke impedance goes up as the freq. going up ( till the resonance of that choke ). One need to look at the whole system not just the first part. 😕
The chokes in the power supply shouldn't be in the signal path correct? 😕
The audio signal taken from the output tube is between plate and cathode right? This would pass through the power supply cap connected to the OPT B+ node and the cathode bypass cap if auto bias. Those caps are the ones you want a low impedance down to 20Hz.
The power supply can also be viewed as a positive feedback loop . Thus the reason for isolation in the power supply so that that positive loop does not effect other parts of the circuit So yes it does effect the signal. System view which is more complex than what many like to teach . Everything maters some just not very much .
- Status
- Not open for further replies.