From kurt–(at)–r.hp.com Wed Dec 6 15:52:13 CST 1995
Article: 4005 of rec.audio.tubes
Path: geraldo.cc.utexas.edu!cs.utexas.edu!uwm.edu!vixen.cso.uiuc.edu!howland.reston.ans.net!nntp.coast.net!col.hp.com!news.dtc.hp.com!canyon.sr.hp.com!kurts
From: kurt–(at)–r.hp.com (Kurt Strain)
Newsgroups: rec.audio.tubes
Subject: Re: Need 5AR4 data
Date: 6 Dec 1995 16:58:51 GMT
Organization: Hewlett Packard Sonoma County
Lines: 59
Message-ID: <4a4i4b$5m--(at)--anyon.sr.hp.com>
References: <49kt0o$s5--(at)--anyon.sr.hp.com> <4a33ad$9b--(at)--arlor.hiwaay.net>
NNTP-Posting-Host: hpsrmta2.sr.hp.com
X-Newsreader: TIN [version 1.2 PL2]
Joe Lowe (jlow–(at)–iwaay.net) wrote:
: TYPICAL OPERATION
: AC plate to plate supply voltage 450 V
: Effective plate supply impedance per plate 160 Ohms
: Average output current 225 Ma
: DC output voltage at input to filter 475 V
: Voltage drop for current of 225 Ma. 17 V
: No problem.
: : What transformer secondary voltage is needed for this? Like maybe 750VCT,
: : or 375Vrms per plate.
: AC input X 1.414 – 17 aprox.
For 420VDC, this formula says that AC input must be (DC output + 17) / 1.414,
or (420+17)/1.414 = 383 Vrms.
Above, where output = 475 VDC, expected input should be (475 + 17) / 1.414,
or 348 Vrms. But, instead, it says we need 450 Vrms. If the effective
plate resistance is 160 ohms and current is .225A, the drop looks like it
will only produce 160 ohms * .225 A = 36 V. Even that is not enough to
account for the large discrepancy between input 450 Vrms and 348 Vrms.
The difference is that series resistance before the input capacitor acts
to average down the voltage output, much like a choke input filter does.
So I modelled it on SPICE to check it out.
With no load, and an AC input of 450 Vrms, the voltage will rise to
1.414*450 = 636 VDC. It doesn’t matter what Rrectifier and Cinput are.
With a 225 mA load, or a resistance load about 2200 ohms, the output voltage
with 160 ohm plate resistance yields not 636 – .225A*160 = 600 VDC, but
actually 450 VDC, as specified, with C = 40 uF. With C = 4000 uF, the
output remains 450 VDC again. The impedance of the tube refuses to allow
a full charge with that current drain.
To get 420 VDC output with 160 ohm rectifier resistance with a 170 mA load,
and 40 uF input capacitance or more, I went back to SPICE to find out it
needed an input of 400 Vrms. An easily obtainable transformer. But it
would be a mistake to buy a 375 Vrms transformer.
A 5U4, on the other hand, has less rectifier resistance, about 50 ohms.
It might be better to use the 375 Vrms transformer for it. Anybody using
5U4’s to replace 5AR4 should expect higher DC voltages in capacitor input
supplies. Choke input supplies make rectifier differences produce much less
impact. That’s because the input choke destroys the full charging capability
anyway, changing it from a peak voltage supply to an average rectified
voltage supply. Series resistance in the rectifier circuit has a similar
effect in ruining the peak voltage supply.
So, I am thankful for the data because I was able to calculate the needed
transformer for it. But I wanted to show how I differ in calculating
the expected output voltage.
Kurt
From kurt–(at)–r.hp.com Thu Dec 7 10:43:55 CST 1995
Article: 4018 of rec.audio.tubes
Path: geraldo.cc.utexas.edu!cs.utexas.edu!news.sprintlink.net!newsfeed.internetmci.com!chi-news.cic.net!nntp.coast.net!col.hp.com!news.dtc.hp.com!canyon.sr.hp.com!kurts
From: kurt–(at)–r.hp.com (Kurt Strain)
Newsgroups: rec.audio.tubes
Subject: Re: Need 5AR4 data
Date: 7 Dec 1995 00:58:16 GMT
Organization: Hewlett Packard Sonoma County
Lines: 69
Message-ID: <4a5e78$n0--(at)--anyon.sr.hp.com>
References: <49kt0o$s5--(at)--anyon.sr.hp.com> <4a33ad$9b--(at)--arlor.hiwaay.net> <4a4i4b$5m--(at)--anyon.sr.hp.com>
NNTP-Posting-Host: hpsrmta2.sr.hp.com
X-Newsreader: TIN [version 1.2 PL2]
Kurt Strain (kurt–(at)–r.hp.com) wrote:
: Joe Lowe (jlow–(at)–iwaay.net) wrote:
: : TYPICAL OPERATION
: : AC plate to plate supply voltage 450 V
: : Effective plate supply impedance per plate 160 Ohms
: : Average output current 225 Ma
: : DC output voltage at input to filter 475 V
: : Voltage drop for current of 225 Ma. 17 V
: : No problem.
: : : What transformer secondary voltage is needed for this? Like maybe 750VCT,
: : : or 375Vrms per plate.
: : AC input X 1.414 – 17 aprox.
: For 420VDC, this formula says that AC input must be (DC output + 17) / 1.414,
: or (420+17)/1.414 = 383 Vrms.
^^^
Sorry, 383 Vrms should read 309 Vrms. Let me change the rest of these
too quickly calculated numbers:
: Above, where output = 475 VDC, expected input should be (475 + 17) / 1.414,
: or 348 Vrms. But, instead, it says we need 450 Vrms. If the effective
: plate resistance is 160 ohms and current is .225A, the drop looks like it
: will only produce 160 ohms * .225 A = 36 V. Even that is not enough to
: account for the large discrepancy between input 450 Vrms and 348 Vrms.
: The difference is that series resistance before the input capacitor acts
: to average down the voltage output, much like a choke input filter does.
: So I modelled it on SPICE to check it out.
: With no load, and an AC input of 450 Vrms, the voltage will rise to
: 1.414*450 = 636 VDC. It doesn’t matter what Rrectifier and Cinput are.
: With a 225 mA load, or a resistance load about 2200 ohms, the output voltage
: with 160 ohm plate resistance yields not 636 – .225A*160 = 600 VDC, but
: actually 450 VDC, as specified, with C = 40 uF. With C = 4000 uF, the
: output remains 450 VDC again. The impedance of the tube refuses to allow
: a full charge with that current drain.
: To get 420 VDC output with 160 ohm rectifier resistance with a 170 mA load,
: and 40 uF input capacitance or more, I went back to SPICE to find out it
: needed an input of 400 Vrms. An easily obtainable transformer. But it
: would be a mistake to buy a 300 Vrms transformer.
Note: Vout is about 1.05*Vrms for both the 420VDC and 475VDC cases.
A choke input supply would produce 1.414*0.637 = 0.900 times Vrms.
This resistive capacitor input situation is somewhere in between.
: A 5U4, on the other hand, has less rectifier resistance, about 50 ohms.
: It might be better to use the 375 Vrms transformer for it. Anybody using
: 5U4’s to replace 5AR4 should expect higher DC voltages in capacitor input
: supplies. Choke input supplies make rectifier differences produce much less
: impact. That’s because the input choke destroys the full charging capability
: anyway, changing it from a peak voltage supply to an average rectified
: voltage supply. Series resistance in the rectifier circuit has a similar
: effect in ruining the peak voltage supply.
: So, I am thankful for the data because I was able to calculate the needed
: transformer for it. But I wanted to show how I differ in calculating
: the expected output voltage.
: Kurt
Kurt, self corrected again.
From kurt–(at)–r.hp.com Wed Dec 13 10:38:20 CST 1995
Article: 4250 of rec.audio.tubes
Path: geraldo.cc.utexas.edu!cs.utexas.edu!bcm.tmc.edu!pendragon!news.msfc.nasa.gov!newsfeed.internetmci.com!news.kei.com!nntp.coast.net!col.hp.com!news.dtc.hp.com!canyon.sr.hp.com!kurts
From: kurt–(at)–r.hp.com (Kurt Strain)
Newsgroups: rec.audio.tubes
Subject: Re: Need 5AR4 data
Date: 12 Dec 1995 19:06:50 GMT
Organization: Hewlett Packard Sonoma County
Lines: 18
Message-ID: <4akjsa$dl--(at)--anyon.sr.hp.com>
References: <49kt0o$s5--(at)--anyon.sr.hp.com> <4a33ad$9b--(at)--arlor.hiwaay.net> <4a4i4b$5m--(at)--anyon.sr.hp.com> <4a5e78$n0--(at)--anyon.sr.hp.com> <4af4at$ce--(at)--anix2.panix.com> <4ahn2d$1m--(at)--anyon.sr.hp.com> <4aiua7$gl--(at)--anix2.panix.com>
NNTP-Posting-Host: hpsrmta2.sr.hp.com
X-Newsreader: TIN [version 1.2 PL2]
Mark Garvin (mgarvi–(at)–anix.com) wrote:
: B+ with no load Under load
:—————– ————
: Silicon rect480 volts 445 v
: 5AR4 (Mullard) 475 426
: 5U4 (RCA) 469 404
Thank you for this data. From this, my simulations indicate that the
load resistance is 1800 ohms, or the current draw is about 0.25A. The
source is 340Vrms. The effective resistance of the transformer is about
90 ohms. The effective static resistance of the 5AR4 is about 50 ohms,
and the effective static resistance of the 5U4 is about 120 ohms. Why
the 5AR4 quotes a resistance spec of 160 ohms, dynamic or otherwise, is
not apparent to me.
Kurt
From cign–(at)–lato.phy.OhioU.Edu Wed Dec 13 10:38:37 CST 1995
Article: 4243 of rec.audio.tubes
Newsgroups: rec.audio.tubes
Path: geraldo.cc.utexas.edu!cs.utexas.edu!bcm.tmc.edu!news.msfc.nasa.gov!newsfeed.internetmci.com!chi-news.cic.net!news.midplains.net!gw2.att.com!oucsboss!cigna
From: cign–(at)–lato.phy.OhioU.Edu (Dave Cigna)
Subject: Re: Need 5AR4 data
X-Nntp-Posting-Host: plato.phy.ohiou.edu (cigna)
Message-ID:
Sender: postmaste–(at)–lato.phy.ohiou.edu (cigna)
X-Nntp-Posting-Date: Tue Dec 12 09:27:38 1995
Organization: Ohio University Physics and Astronomy
References: <49kt0o$s5--(at)--anyon.sr.hp.com> <4a5e78$n0--(at)--anyon.sr.hp.com> <4af4at$ce--(at)--anix2.panix.com> <4ahn2d$1m--(at)--anyon.sr.hp.com>
Date: Tue, 12 Dec 1995 14:27:39 GMT
Lines: 57
In article <4ahn2d$1m--(at)--anyon.sr.hp.com>, Kurt Strain
>Mark Garvin (mgarvi–(at)–anix.com) wrote:
>: In <4a5e78$n0--(at)--anyon.sr.hp.com> kurt–(at)–r.hp.com (Kurt Strain) writes:
>
>: >: A 5U4, on the other hand, has less rectifier resistance, about 50 ohms.
>: >: It might be better to use the 375 Vrms transformer for it. Anybody using
>: >: 5U4’s to replace 5AR4 should expect higher DC voltages in capacitor input
>: >: supplies.
>
>: Hi Kurt, In comparing a 5U4 to 5AR4, I’ve found that the 5U4 drops
>: considerably MORE voltage than the 5AR4.
>
>Then either the 5U4 datasheet is wrong in stating their plate impedance
>or the 5AR4 is not 160 ohms plate resistance. Could someone please double
>check? I’m talking about the conditions for 420VDC, 170mA operation, and
>they need to be apples to apples comparison. Only plate impedance at
>given voltage and current effects the output for a given input capacitor.
>Something must give. I wish I had the datasheet for the 5AR4, but I
>don’t. It must be a late model, or one RCA never made. Thanks.
Hi Kurt,
My Sylvania tube book mentions a ‘limiting resistor’ of 125 ohms for 430VDC
operation. Evidently, this is a series resistor (!) that they suggest
adding. Could this be what your data sheet is talking about? I
misinterpreted it the first time I read it.
Anyway, here’s most of what’s in the Sylvania book:
Design center ratings:
heater current; 1.9 amps
peak inverse voltage: 1500 V
peak plate current: 750 mA
AC plate supply (max RMS each plate): 550 V
steady state DC output current: 250 mA
condenser input to filter (max): 60 uF
Typical operations:
Supply (RMS per plate): 300 350 400 450 500 550
DC output current (mA): 250 250 250 250 250 200
DC output voltage: 330 380 430 480 560 640
Limiting resistor 75 100 125 150 175 200
(per plate)
For a real-world example: I installed a Sovtek 5AR4 in a reissue
Bassman. I measured the transformer sec. at 345 VAC (~115VAC line).
With the amp idling (75-80 mA) and 50uF input capacitance there
was just about 460 VDC at the first filter. This amp has been
running fine for about six months, and I know that the owner likes
to play loud whenever he can.
— Dave Cigna
From stephen.delf–(at)–elcom.gen.nz Thu Dec 14 11:50:33 CST 1995
Article: 4307 of rec.audio.tubes
Path: geraldo.cc.utexas.edu!cs.utexas.edu!math.ohio-state.edu!uwm.edu!newsfeed.internetmci.com!in1.uu.net!comp.vuw.ac.nz!welcom!stephen.delft
From: stephen.delf–(at)–elcom.gen.nz (Stephen Delft)
Newsgroups: rec.audio.tubes
Subject: Re: Need 5AR4 data
Date: Thu, 14 Dec 1995 07:33:00 GMT
Message-ID: <951214234537298--(at)--elcom.gen.nz>
Organization: WELCOM BBS
Distribution: world
References: <4aiua7$gl--(at)--anix2.panix.com> <49kt0o$s5--(at)--anyon.sr.hp.com> <4a33ad$9b--(at)--arlor.
Lines: 44
MG> First cap is 30 uf.
MG>B+ with no load Under load
MG> —————– ————
MG>Silicon rect480 volts 445 v
MG>5AR4 (Mullard) 475 426
MG>5U4 (RCA) 469 404
MG>Anyone disagree?
Hi Mark, the metal platers have some kind of problem with the bath, so
I have a few days break from those amps
I got similar figures for the different rectifiers:
OPTX = 372-0-372 volts AC, with an additional 33 Ohm ballast in each
plate lead. Can’t remember the first cap, but probably 2 x 47uF in
series, so take it as 23uF
At continuous DC current of 100mA …
Two BYW96E Sil diodes.(Phil.1kV.3A) 479 V DC
GZ34/5AR4 (Mullard) 464 V DC
2 x 5Y3 wired in parallel 458 V DC
1 x 5Y3 440 V DC
5U4GB433 V DC
…so if we take the diode drop as small ( about 0.5 Volt)
at 100mA DC, 5AR4 drops about 15 V
Two x 5Y3 21 V
Single 5Y3 39 V
5U4 46 V
I didn’t measure a 5U4, but from memory its static voltage drop is
is somewhere between 5AR4 and 5Y3.
BTW, If you are building from scratch, and can provide two sockets
and the (total) 3 Amp at 5 Volt fil. current, then TWO x old Russian
5Y3’s (wrongly marked as GZ34) do a fine job in a 40 Watt guitar amp,
and are rated low enough to expect a long life. Sounds brighter than
5V4 or 5U4; similar to 5AR4, but a bit more compressed.
Cheers, Stephen.
