pucho812
Well-known member
what is a general rule here regarding decoupling capacitors in vacuum designs. I can't seem to find much in terms of google on the subject.
decoupling capacitors in vacuum tube designs
- Thread starter pucho812
- Start date
Help Support GroupDIY Audio Forum:
I am not a tube guy, but why would there be different rules?
JR
JR
pucho812
Well-known member
Well I am curious in particular to d.c. Heaters and if there is need to decouple
L´Andratté
Well-known member
bluebird said:
decoupling vs. coupling misunderstanding?
john12ax7
Well-known member
In guitar amps 10uF per preamp tube is pretty typical
Decoupling the cathode does tend to help reduce induced noise from the heater .
You'll also get a handfull more db gain with decoupling ,the exact amount depends on the tube and circuit values.
Bass response will also be different depending on your value of decoupling ,generally though the decoupling cap is made large enough to have negligeable effect compared to the coupling cap.
In guitar amps you sometimes find quite small decoupling on the inputs , I guess this will give a step to the bass response, more often than not a value of tens or hundreds of uf is what you find.
Push pull Output stages with cathode bias and class a/b or above require decoupling to function correctly .
I did find a simple comparison chart for an ecc83/12ax7 with and without decoupling before ,try your search engine with someting like '12ax7 unbypassed vs bypassed' you should find it .
You'll also get a handfull more db gain with decoupling ,the exact amount depends on the tube and circuit values.
Bass response will also be different depending on your value of decoupling ,generally though the decoupling cap is made large enough to have negligeable effect compared to the coupling cap.
In guitar amps you sometimes find quite small decoupling on the inputs , I guess this will give a step to the bass response, more often than not a value of tens or hundreds of uf is what you find.
Push pull Output stages with cathode bias and class a/b or above require decoupling to function correctly .
I did find a simple comparison chart for an ecc83/12ax7 with and without decoupling before ,try your search engine with someting like '12ax7 unbypassed vs bypassed' you should find it .
bluebird
Well-known member
L´Andratté said:
Doh! sorry Pucho, yeah I was thinking "coupling" between stages. never mind, post removed.
buildafriend
Well-known member
I have a box designed just for toying with this. It's a variable resistor in parallel with a few selectable capacitor values. You could make one and observe what the results are. It's an Rk and Ck box
If you're talking about coupling caps it blocks the DC from the power supply from going to the next stage and sets the LF cutoff frequency
If you're talking about coupling caps it blocks the DC from the power supply from going to the next stage and sets the LF cutoff frequency
pucho812
Well-known member
correct. But was talking decoupling caps.buildafriend said:
buildafriend
Well-known member
Cool. For toying with small parts of circuits like this it's nice to have some C's and R's on stepped switches to see what the practical effects are at the flick of a switch. You can put clip leads or wires coming off of the switch to make it easy to put in place of the fixed value that used to be there and throw a volt or ammeter in its path to keep an eye on I or V changes as the result while running f sweeps and watching THD changes
Diabolical Artificer
Active member
- Joined
- Aug 12, 2014
- Messages
- 32
To find the right value of coupling cap use 1/(2 times pi times f times r) r being your grid resistor. So for most stages f would be about 15hz abouts. In a lot of old amps they use 10n with 1meg resistors, it's a good rule of thumb. See - http://www.valvewizard.co.uk/gainstage.html well worth a read.
As regards the DC voltage rating 630v is a common value and covers most amp topology's.
Andy.
As regards the DC voltage rating 630v is a common value and covers most amp topology's.
Andy.
ruffrecords
Well-known member
A lot depends on the particular tube circuit. Many simple tube topologies (and for that matter simple semiconductor ones) have a fairly poor power supply rejection ratio (PSRR), some as little as 6dB. This means you do need more power supply decoupling than an op amp for example.
If you want power supply hum to be below 100dBu this is less than 1uV at the output. If your PSSR is only 6dB you need to make sure the decoupling drops it to around 2uV at the supply.
Successive RC networks are more effective than a single RC network of the same total RC.
Cheers
Ian
If you want power supply hum to be below 100dBu this is less than 1uV at the output. If your PSSR is only 6dB you need to make sure the decoupling drops it to around 2uV at the supply.
Successive RC networks are more effective than a single RC network of the same total RC.
Cheers
Ian
There are some solid state designs with poor PSRR too... so same calculus with perhaps different variables.
Indeed a fair generalization that tube circuits require more PS decoupling than IC circuits.
JR
Indeed a fair generalization that tube circuits require more PS decoupling than IC circuits.
JR
Audio1Man
Well-known member
pucho812 said:
There is no magic value for decoupling, only good engineering practices and verification.
Power supply decoupling is needed to reduce the HUM, RIPPLE & or SIGNAL from detreating the performance of the product. In the old days a scope was used to check for the problems and with today’s tools we have improved methods. Electrolytic capacitors have typical tolerances that are wide +20%, -50% is not uncommon. They sometime only work @ low frequencies and require additional HF bypassing.
I verify my designs or repairs with a scope and my Audio Precision analyzer to check for noise /signals on the power supply nodes. The Scope has high voltage input tolerance and has a 1 volt output that can connect to and analyzer for Analog or FFT measurements for debugging.
I also use a signal injection method to check the product for correct functions. Using a signal source I drive a current into the node and see what comes out of the product.
Duke
+1 I used signal injection on my bench back in the 70s when I was trying to figure this stuff out.Audio1Man said:
The signal source could be any sine wave generator that can drive a few hundred ohms. Using a capacitor in series with a resistor you could drive a few mA AC signal into DC rails. Varying the sine wave frequency could reveal how stiff or not the rail was at higher/lower frequencies. You could even inject some audio AC into a rail to see where and how much comes out elsewhere. :
JR
buildafriend
Well-known member
JohnRoberts said:
My first sine wave generator was a $30 kit from eBay with two banana jacks out that worked great. It's nice to have an accurate display that tells you what you are actually putting out but I used to just measure it with a DMM or my Tek scope. It was unbalanced out but usually did the trick.. I guess you could use your phone now. They have free apps for sine waves but I'm not sure how well it works or if you could damage your HP jack with some DC.
This might work if you're afraid to use your phone or DAW on a faulty DUT
https://www.ebay.com/itm/New-XR2206-Signal-Generator-Sine-Triangle-Square-Wave-Frequency-Adjustable-W1Y4/222236383728?hash=item33be5219f0:g:mUoAAOSwOdpXx9Tl
All the best,
Good practice at tube decoupling is having cap + connection very close to decoupled tube's anode, - close to point where cathode joins ground. Having all tube's ground conections at one point, short fat ground bus to other tubes/points like first helps with noise. You don't get higher current ripple into sensitive inputs because output from driver goes directly to common ground at psu/IEC.
I use 22-47u depending on tube, have to try less because supplies are extremely well filtered and stabilized by zeners.
Some old schematics show very little, 8u or even nothing.
i took two examples; Gates SA70 and W.E.141-A, nones has decoupling. EMRR made modern "mash-up" of this two with triode wired 6J7 and .1u decoupling, 6SN7 as used by W.E. 10u. IIRC European gear tends to use more tube decoupling and i don't really know why.
I use 22-47u depending on tube, have to try less because supplies are extremely well filtered and stabilized by zeners.
Some old schematics show very little, 8u or even nothing.
i took two examples; Gates SA70 and W.E.141-A, nones has decoupling. EMRR made modern "mash-up" of this two with triode wired 6J7 and .1u decoupling, 6SN7 as used by W.E. 10u. IIRC European gear tends to use more tube decoupling and i don't really know why.
moamps
Well-known member
A lot of info about tube designs you can find in Morgan Jones Valve Amplifiers book.pucho812 said:
You can download it for free (third edition) from web.
moamps
Well-known member
What's the purpose of C7.2 and C7.3 in WE141A design?My3gger said:
