Anyone ever measure the A470's impedance?

knowledge base for the classic Dynaco ST70

Anyone ever measure the A470's impedance?

Postby Shannon Parks » Mon May 23, 2011 6:12 am

I tested my plastic leaded, Japanese A470 output transformers last night and measured 5.9K to 8 ohms. Bandwidth was <2Hz to 63kHz.

I measured them with a signal generator (Wavetek FG for the sub 10Hz testing, Boonton 1120 for the rest) connected to the plate connections (blue and blue-white) and an 8 ohm load on the black to orange wire, and monitored the actual input and output voltages. Came up with impedance: square of the ratio (i.e. input/output) * 8 ohms.

Does this sound like a proper test? I'm planning on making a 1625 ST70 which I'm dubbing the JAN70. I was thinking the 4.3K impedance would be too low for 1625s, so I was going to use the 4 ohm tap (for 8 ohm loads) to get a better match. At almost 6K, this seems much better. Obviously I will test at the top level and compare there, but I'm trying to do a better job of evaluating iron.

Shannon
User avatar
Shannon Parks
Site Admin
 
Posts: 3764
Joined: Tue Mar 18, 2003 5:40 pm
Location: Poulsbo, Washington

Postby EWBrown » Tue May 24, 2011 12:26 pm

Hmmm, I always presumed that an A470 had a 4.3K primary. :/

I'll have to dig up my single "spare" A470, test it with low voltage 60 Hz(15 to 20VAC) , and see what I get...

As far as I know, 1625s can be treated just like 807s, (they are nearly identical where it really counts), 6L6GCs and KT66s, as far as operating voltages, currents and OPT primary impedances.

These can vary from 4K A-A to 9K A-A, depending on voltage & current on the output tubes. [:)

/ed B
Last edited by EWBrown on Wed Nov 05, 2014 8:33 pm, edited 2 times in total.
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Postby Geek » Tue May 24, 2011 4:02 pm

Yeah, I've measured 4.25-4.4 on various A470's (???)

It's the DCR that's inconsistent as heck on Dynaco OPT's between manufacturers and runs =:o

Cheers!
-= Gregg =-
Fine wine comes in glass bottles, not plastic sacks. Therefore the finer electrons are also found in glass bottles.
User avatar
Geek
KT88
 
Posts: 3585
Joined: Sun Oct 21, 2007 3:01 am
Location: Chilliwack, British Columbia

Postby Shannon Parks » Wed May 25, 2011 4:56 am

Let me test a "cloth lead" A470 while I'm drinking my coffee. (c)

Shannon
User avatar
Shannon Parks
Site Admin
 
Posts: 3764
Joined: Tue Mar 18, 2003 5:40 pm
Location: Poulsbo, Washington

Postby Shannon Parks » Wed May 25, 2011 6:23 am

Plastic Lead A470
5663 ohms to 8 ohm. HF -3dB: 69kHz.
(8.86V in with .333V out)

Cloth Lead A470
5741 ohms to 8 ohm. HF -3dB: 66kHz.
(8.76V in with .327V out)

Hammond 1650N
4482 ohms to 8 ohm. HF -3dB: 117kHz.
(9.09V in with .384V out)

10V 1kHz input at primary which was remeasured, with 8 ohm resistive load on output. I can fetch stamp codes later for those interested. The Hammond does test as an impressive piece of iron. No wonder my Ikes always sounded fantastic.

FWIW, it certainly seems as though the cloth lead and plastic lead trannies are functionally identical. My cloth leaded ones are definitely early - like around 1960 - due to the style of the original kit.

Shannon
User avatar
Shannon Parks
Site Admin
 
Posts: 3764
Joined: Tue Mar 18, 2003 5:40 pm
Location: Poulsbo, Washington

Postby Geek » Wed May 25, 2011 7:38 pm

Hmmm, I'll have to find my paperwork. Maybe I tested with the wrong secondary.....

Cheers!
-= Gregg =-
Fine wine comes in glass bottles, not plastic sacks. Therefore the finer electrons are also found in glass bottles.
User avatar
Geek
KT88
 
Posts: 3585
Joined: Sun Oct 21, 2007 3:01 am
Location: Chilliwack, British Columbia

Postby dcriner » Thu May 26, 2011 4:51 pm

Input impedance of an audio output xfmr is a function only of the turns ratio and the output load (speaker) impedance:

Input impedance = (turns ratio)^2 x speaker impedance

The impedance of an audio xfmr, by itself, unconnected to a load, is meaningless.

The speaker impedance (and therefore the xfmr input impedance) is usually specified at 400 Hz. At other frequences, it will be significantly different. DC resistance also is meaningless.
Doug Criner
dcriner
KT88
 
Posts: 309
Joined: Mon Oct 23, 2006 4:19 pm
Location: Illinois

Postby dcgillespie » Fri May 27, 2011 8:19 pm

Shannon -- This is very interesting. If true, it would certainly be a significant deviation from published specifications, which always implied that the A470 was just a little brother to the A430 series, with all things basically equal except for the power rating. Two things you might check:

1. Try the test using the full 16 ohm secondary. It is well known that the taps on some transformers are not as accurate in maintaining the stated primary impedance as the full secondary produces. I recall testing some of these transformers many, many moons ago, and do not recall any deviation like you are finding -- but then I always test using the full secondary, so that may be the difference.

2. It would be interesting to see what percent of the winding your screen taps represent. The A430 series and A470 have 33% taps on the primary winding, which should remain consistent regardless of how well the stated secondary impedance maintains the designated primary impedance.

Thanks for supplying the data.

Dave
dcgillespie
KT88
 
Posts: 399
Joined: Sun Mar 08, 2009 5:34 am
Location: Ball Ground, GA

A470 impedance and others

Postby ronaldw441 » Sat May 28, 2011 12:54 am

Please describe the best way to measure the bandwidth of an output transformer when out of circuit. I would like to compare a Hammond 1650N to an Edcor CXPP60-MS-4.2K.

Here are some results from several transformers I've tested. Unfortunately I've been inconsistent. Some were measured using a variac, others with a signal generator. None have been loaded and none at 400hz. I'll have to go back & try dcriner's method.

All results at the 8 ohm tap unless otherwise specified.

Conrad Johnson MV-45 (Northlake Engineering C126-110A)
Transformer 1 – 3852 ohms
Transformer 2 – 3852 ohms

Conrad Johnson MV-50 (Triad-Utrad 17-8742)
Transformer 1 – 5026 ohms
Transformer 2 – 5019 ohmns

Dynaco ST-70 (Dynaco A-470 cloth leads)
At 8 ohm taps
Transformer 1 – 4788 ohms
Transformer 2 – 4786 ohms

At 16 ohm taps
Transformer 1 – 4286 ohms
Transformer 2 – 4284 ohms

Dynaco Mark III (Dynaco A-431)
At 8 ohm taps
Transformer 1 – 4799 ohms
Transformer 2 – 4790 ohms

At 16 ohm taps
Transformer 1 – 4301 ohms
Transformer 2 – 4292 ohms

Yaqin MC-10L
Transformer 1 – 4303 ohms
Transformer 2 – 4302 ohms


VAA MkIII (EL34 UL)
Transformer 1 – 4118 ohms
Transformer 2 – 4108 ohms

Hammond 1650N (specified as 4300 ohms)
Transformer 1 – 4258 ohms
Transformer 2 – 4246 ohms
Transformer 3 – 4258 ohms
Transformer 4 – 4282 ohms

Edcor CSPP60-MS-4.2K (specified as 4200 ohms)
Transformer 1 – 4184 ohms
Transformer 2 – 4150 ohms

Jolida SJ-502A (SG- 50A [Shuguang?] transformers)
Transformer 1 – 4122 ohms
Transformer 2 – 4086 ohms

Lite T-62 (Chinese Marantz 8 clone)
Transformer 1 – 4395 ohms
Transformer 2 – 4398 ohms
ronaldw441
 
Posts: 4
Joined: Thu May 26, 2011 8:11 pm
Location: Broomfield, CO USA

Postby Shannon Parks » Sat May 28, 2011 6:06 am

I tested these as per my earlier post: a 10V, 1kHz input which was remeasured (due to loading on the primary) using the Boonton 1120 which was also used to measure the output. FWIW, I also tested my loads with a 4-wire measurement: 15.56 ohms, 8.06 ohms, and 4.02 ohms.

It seems to me that input frequency really shouldn't matter for a hifi output - we are expecting 0dB deviation from 100Hz to 10kHz, right? But choosing a frequency that one can reasonably measure with your bench tools - like even a cheap DMM - seems the first importance. 400Hz to 1kHz seems like the standard range for that type of device.

Plastic A470
16 ohm -- 8.59V to .484V -- 4901 ohms
8 ohm -- 8.76V to .3285V -- 5732 ohms
4 ohm -- 8.67V to .2354V -- 5453 ohms

Cloth A470
16 ohm -- 8.58V to .485V -- 4870 ohms
8 ohm -- 8.75V to .3286V -- 5715 ohms
4 ohm -- 8.67V to .2356V -- 5444 ohms

Hammond 1650N
8 ohm -- 8.58V to .356V -- 4682 ohms
4 ohm -- 8.58V to .2518V -- 4668 ohms
User avatar
Shannon Parks
Site Admin
 
Posts: 3764
Joined: Tue Mar 18, 2003 5:40 pm
Location: Poulsbo, Washington

Postby Shannon Parks » Sat May 28, 2011 6:11 am

dcgillespie wrote:2. It would be interesting to see what percent of the winding your screen taps represent. The A430 series and A470 have 33% taps on the primary winding, which should remain consistent regardless of how well the stated secondary impedance maintains the designated primary impedance.


Pardon my ignorance but how do I make this measurement?

Shannon
User avatar
Shannon Parks
Site Admin
 
Posts: 3764
Joined: Tue Mar 18, 2003 5:40 pm
Location: Poulsbo, Washington

Postby Geek » Sat May 28, 2011 6:23 am

Ah! maybe that's why I got published specs... I used 60Hz.

Cheers!
-= Gregg =-
Fine wine comes in glass bottles, not plastic sacks. Therefore the finer electrons are also found in glass bottles.
User avatar
Geek
KT88
 
Posts: 3585
Joined: Sun Oct 21, 2007 3:01 am
Location: Chilliwack, British Columbia

Re: A470 impedance and others

Postby Shannon Parks » Sat May 28, 2011 6:27 am

ronaldw441 wrote:Please describe the best way to measure the bandwidth of an output transformer when out of circuit. I would like to compare a Hammond 1650N to an Edcor CXPP60-MS-4.2K.

Here are some results from several transformers I've tested. Unfortunately I've been inconsistent. Some were measured using a variac, others with a signal generator. None have been loaded and none at 400hz. I'll have to go back & try dcriner's method.


I'm still learning this too, but here is my stab at it:

1) Go ahead and yes, re-test the transformers loaded. Test them on the tap you will be using with a resistive load that most closely matches your intended speaker load. Note impedance.

2) Increase frequency to find the HF -3dB. Take the nominal output - say you are measuring 300mV - and multiply it by .707. Increase the frequency until your output drops to 212mV. That's your HF cutoff. Higher is better, and will permit higher levels of feedback and stability.

3) Difficult step you might omit, since values will normally be under 5Hz and that gets difficult to read, even if your function generator can go down that low. Decrease frequency to find the LF -3dB. Take the nominal output - say you are measuring 300mV - and multiply it by .707. Decrease the frequency until your output drops to 212mV. That's your LF cutoff. Lower is better, and will permit higher levels of feedback and stability. Maybe look for the -1dB at LF instead, since that will be much easier. Multiply by .89.

4) Shoot a 10kHz square wave through the tranny and monitor the loaded output on your oscilloscope. Less ringing is better. It is possible you might have a tranny that has a higher HF cutoff than another, but has more ringing. I would then select the transformer that rings less as the superior transformer.

These tests don't test saturation/output power capability. For that, just go with the boat anchor test.
(lol)

I'm very interested in your Edcor test - please let us know your results.

Shannon
User avatar
Shannon Parks
Site Admin
 
Posts: 3764
Joined: Tue Mar 18, 2003 5:40 pm
Location: Poulsbo, Washington

Postby dcgillespie » Sat May 28, 2011 7:11 am

Shannon, I have always used a variac for my measurements, which makes measuring easier, and provides a guaranteed low impedance drive for the tests.

To measure the screen tap percentage, if (for example), 100.0 vac is applied between the two plate leads, then the voltage between the two screen leads can be measured to determine the percentage. If you measure 33.0 vac in this instance, that would be a screen tap percentage of 33%.

FWIW, I always measure the turns ratio of a transformer with no load connected. This could be a possible variation as well.

Dave
dcgillespie
KT88
 
Posts: 399
Joined: Sun Mar 08, 2009 5:34 am
Location: Ball Ground, GA

Postby Shannon Parks » Sat May 28, 2011 7:45 am

Thanks, Dave!

Shannon
User avatar
Shannon Parks
Site Admin
 
Posts: 3764
Joined: Tue Mar 18, 2003 5:40 pm
Location: Poulsbo, Washington

Next

Return to st70

Who is online

Users browsing this forum: Google [Bot] and 20 guests

cron