UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Audio (record players, hi-fi etc)

Notices

Vintage Audio (record players, hi-fi etc) Amplifiers, speakers, gramophones and other audio equipment.

Reply
 
Thread Tools
Old 10th Nov 2023, 5:01 pm   #1
Impecunious
Triode
 
Join Date: Oct 2011
Location: Birmingham, West Midlands, UK.
Posts: 28
Default Truvox TSA 100 Amp and OC81Z

One channel has lower gain than the other (about half peak to peak voltage at output). I have taken out the OC81Z (VT7 on diagram) and its Hfe is 10 ( measured on MK328) which to me seems miserly. In circuit, the voltages on its leads are fine, with base being about 200mV different from emmiter. The voltages around the circuit are pretty much as in the diagram.

The AF118 does get moderately warm, something I would not expect. The AF118 in the stage working properly is cold.

A few questions;
1) would the low Hfe of the OC81Z account for the output stage's overall low gain?
2) I appreciate that the OC81Z will have at least 25V across it before there is any signal, and if I were to replace it, how critical is it that the Vce of the replacement is 40V, like the OC81Z? It has few equivalents, only ASY48.
3) Could I put in an AC128 or an OC81 (Vce 20V) and hope that it doesn't notice that its getting a few more volts than its manufacturer intended.

Thanks David
Attached Files
File Type: pdf TSA-100-1.pdf (1.06 MB, 115 views)

Last edited by Impecunious; 10th Nov 2023 at 5:04 pm. Reason: forgot attachment
Impecunious is offline   Reply With Quote
Old 10th Nov 2023, 7:02 pm   #2
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,887
Default Re: Truvox TSA 100 Amp and OC81Z

From memory the Z-rated transistors were selected to handle a higher emitter-collector voltage than the generic parts.

This would be supported by the use of an AF118 - which I would more associate with first-generation transistorised TV deflection stages.
__________________
I'm the Operator of my Pocket Calculator. -Kraftwerk.
G6Tanuki is offline   Reply With Quote
Old 10th Nov 2023, 7:27 pm   #3
Jez1234
Banned
 
Join Date: Mar 2021
Location: Morpeth, Northumberland, UK.
Posts: 936
Default Re: Truvox TSA 100 Amp and OC81Z

Assuming you've checked out the other stages and the fault is definitely in the power amp then I'd check the 50uF emitter decoupling cap on the AF118 and the 12.5uF bootstrap cap.
Jez1234 is offline   Reply With Quote
Old 11th Nov 2023, 11:02 am   #4
SteveCG
Nonode
 
Join Date: Jun 2009
Location: Herefordshire, UK.
Posts: 2,494
Default Re: Truvox TSA 100 Amp and OC81Z

As G6Tanuki says the Z almost certainly means specially selected for higher Emitter to Collector voltage working. The book on Leak amplifiers comments the same about some of the devices used by Leak in their early transistor amplifiers.
SteveCG is offline   Reply With Quote
Old 15th Nov 2023, 1:12 pm   #5
Impecunious
Triode
 
Join Date: Oct 2011
Location: Birmingham, West Midlands, UK.
Posts: 28
Default Re: Truvox TSA 100 Amp and OC81Z

Thanks for suggestions, an update.
1) OC81Z, Hfe 10 (VT7) on diagram was replaced with OC84D with Hfe 65. This made no difference, there is still a discrepancy between the gain of the left and right power stages.
I've injected identical signal to base of VT5 VT15 and the difference is still apparent at the output, so it cannot be attributable to the previous stages.

2) Electrolytics C20, C21, C23, C25 all removed, tested and all found to be good, but having removed them, I replaced with new ones anyway. Still no difference to gain of the left and right power stages.

3 Tested VT6, VT7, VT8, VT9 VT10 and all good. Mid rail voltage is 25V as it should be, so confirms that output pair are OK.

4) I have put gain difference to one side and have tackled noise. Both stages extremely noisy and have easily cured this by changing base bias resistors, R31, R32 on VT5 and the collector and emmiter load resistors, and all their equivalents on the other channel. From being dreadfully noisy, the left and right power stages are now virtually noise free. The originals were carbon composition.

5) Tone control stage very noisy and will also have its carbon compositions replaced. As well as bias resistors, I'll also replace collector and emmiter resistors.

This must be a later amp as VT1,2,3,4 and 5 are not OC germanium types, but 2N4058 Si.

I sometimes wonder about the wisdom of tackling amps with Ge and carbon comp resistors, but I will probably pursue. Thanks for suggetstions so far.
David
Attached Thumbnails
Click image for larger version

Name:	P1100509.jpg
Views:	143
Size:	135.8 KB
ID:	287908  

Last edited by Impecunious; 15th Nov 2023 at 1:21 pm.
Impecunious is offline   Reply With Quote
Old 15th Nov 2023, 3:54 pm   #6
DMcMahon
Dekatron
 
DMcMahon's Avatar
 
Join Date: Apr 2019
Location: Worthing, West Sussex, UK.
Posts: 6,556
Default Re: Truvox TSA 100 Amp and OC81Z

You are doing well David, good logical fault finding/testing.

I guess this is a standalone amp not part of a tape machine so really should be moved by the moderators to the Vintage Audio section.

David
DMcMahon is online now   Reply With Quote
Old 16th Nov 2023, 2:22 am   #7
Impecunious
Triode
 
Join Date: Oct 2011
Location: Birmingham, West Midlands, UK.
Posts: 28
Default Re: Truvox TSA 100 Amp and OC81Z

Ah yes, I now notice that I'm in the wrong place, and not for the first time in my life. The TSA is a 10W per channel standalone amp. Mods, please move me.
Ta D
Impecunious is offline   Reply With Quote
Old 16th Nov 2023, 8:57 am   #8
Station X
Moderator
 
Station X's Avatar
 
Join Date: Jan 2003
Location: Ipswich, Suffolk, IP4, UK.
Posts: 21,035
Default Re: Truvox TSA 100 Amp and OC81Z

Thread moved.
__________________
Graham. Forum Moderator

Reach for your meter before you reach for your soldering iron.
Station X is online now   Reply With Quote
Old 17th Nov 2023, 2:29 pm   #9
dave cox
Nonode
 
dave cox's Avatar
 
Join Date: Jan 2009
Location: Bristol, UK.
Posts: 2,043
Default Re: Truvox TSA 100 Amp and OC81Z

Does VR52 adjust the gain ?

dc
dave cox is offline   Reply With Quote
Old 17th Nov 2023, 4:01 pm   #10
DMcMahon
Dekatron
 
DMcMahon's Avatar
 
Join Date: Apr 2019
Location: Worthing, West Sussex, UK.
Posts: 6,556
Default Re: Truvox TSA 100 Amp and OC81Z

Do the transistor voltages on the output stage measure similar between left and right channels and are the voltages similar to the values shown on the schematic ?

David
DMcMahon is online now   Reply With Quote
Old 20th Nov 2023, 9:27 pm   #11
Impecunious
Triode
 
Join Date: Oct 2011
Location: Birmingham, West Midlands, UK.
Posts: 28
Default Re: Truvox TSA 100 Amp and OC81Z

Do the transistor voltages on the output stage measure similar between left and right channels and are the voltages similar to the values shown on the schematic ?

Yes

Voltages in the L and R Output stages are pretty similar and are as the diagram. I'm assuming that mid-rail voltage should be half supply, which it is.
This mid rail is altered by VR42, which also changes the quiescent current. Voltages on the output AD149s are all good, with the bases being about 150mV above the emitter. The same can be said of the OC81 and AC 127 drivers. The AF118 also has its base about 0.2V above that of its emitter. This suggests to me that the semiconductors are all good. The 2N4058, which is VT5 has its base 0.6V above emitter, which would make sense as it’s Si.

One thing I have noticed is that the quiescent current ( calculated by measuring the voltage across the 2.2 Ohm emitter resistors) for the Right channel, ( the one with lower gain ) is 6mA, whilst the left channel quiescent current is 3.9mA. Those are with mid rail set at half supply. I can of course change quiescent current, but that changes the mid rail voltage. Those current are accurate as I replaced all the emitter resistors with new ones and even checked that they were 2.2 Ohm before they went in. (Old ones I took out were, 2.7, 3.8, 7.3 and 13.0 !)

The circuit has 0.03 V across emitter resistors R49 and R50, and this gives current of 15mA. (Diagram has them as 2.0 Ohm, but in reality they were 2.2)

Should I try setting VR42 to the spec of 15mA current and just forget about mid rail voltage being half supply.

VR52 adjusts gain only minimally, not enough to correct for discrepancy.
I’m still stumped.
Anyway I replaced all the resistors in the tone control stage and what once was a roaring sea of noise on both channels is now just a whisper. So success there.
Again, thanks for help so far.
Impecunious is offline   Reply With Quote
Old 20th Nov 2023, 11:15 pm   #12
frsimen
Heptode
 
Join Date: Jan 2011
Location: Croydon, London, UK.
Posts: 758
Default Re: Truvox TSA 100 Amp and OC81Z

With VR52 not making much difference to the gain, check that R38 is in good health. If it has gone high in value, the gain will be reduced and the range of adjustment from VR52 will be reduced.

C50, the bootstrap capacitor, will also have a big influence on the gain. You haven't said whether you have checked that but it was mentioned by Jez1234 earlier in the thread.

The voltage gain from the base of VT15 to the output should adjust from about 22 to 122, if all is working well.

Check also R104 on the other channel. If that has gone high, that channel's gain will be higher than it should without adjustment of VR105.

Paula
frsimen is offline   Reply With Quote
Old 20th Nov 2023, 11:45 pm   #13
Impecunious
Triode
 
Join Date: Oct 2011
Location: Birmingham, West Midlands, UK.
Posts: 28
Default Re: Truvox TSA 100 Amp and OC81Z

First of all, many thanks with your persistance with this... almost as great as mine !

With VR52 not making much difference to the gain, check that R38 is in good health. If it has gone high in value, the gain will be reduced and the range of adjustment from VR52 will be reduced.

Have done so, its around the 22 it should be. tested in circuit, both ways around so that no stray volts could upset meter.


C50, the bootstrap capacitor, will also have a big influence on the gain. You haven't said whether you have checked that but it was mentioned by Jez1234 earlier in the thread.

All electrolytics (C20, C50, C22, C23, C25, ) replaced. C26, 1000uF tested to be spot on.

The voltage gain from the base of VT15 to the output should adjust from about 22 to 122, if all is working well.

I'll go and take some measurements of voltage gain for each channel. You say it should be adjustable.... via VR52 ?

Check also R104 on the other channel. If that has gone high, that channel's gain will be higher than it should without adjustment of VR105.

I was wondering if I were chasing the wrong channel and that the fault was too much gain in one stage rather than too little in the other. I'm sure I've checked all the values of all the components except the 600pf caps, C24, C27, C51, C54

One last thing, I've been trying at all frequencies and the difference in gain starts to become less noticable above about 12KHz. By 30k ( yes I know the amp is only designed to 20k) the gains of L and R are both low, but the same. Makes me think capacitative...... somewhere.


Thanks again for help and hints.


Paula
Impecunious is offline   Reply With Quote
Old 21st Nov 2023, 12:02 am   #14
frsimen
Heptode
 
Join Date: Jan 2011
Location: Croydon, London, UK.
Posts: 758
Default Re: Truvox TSA 100 Amp and OC81Z

Yes, the gain should adjust with VR52 on the "bad" channel or VR105 on the other channel.

Paula
frsimen is offline   Reply With Quote
Old 21st Nov 2023, 7:32 pm   #15
Impecunious
Triode
 
Join Date: Oct 2011
Location: Birmingham, West Midlands, UK.
Posts: 28
Default Re: Truvox TSA 100 Amp and OC81Z

HOO ****** Ray
Sorted, at last and thanks for all your help.
The "bad" channel (with low gain) was indeed good and the baddun was overamplifying due to the pot VR52 having an openish track of 32k rather than the 2k it should have had. Anyway, found a preset to put in and both channels track beautifully with identical gain at all frequencies. So that poor channel that had been subject to nearly every component being lifted and tested and in many cases replaced, was in fact never the guilty party. Oh how wronged it must feel.

After I realised that most resistors were noisy I replaced nearly all and the attached pic gives an idea of what's come out of the earlier stages as well. The emmitter resistors aren't there as I'd tossed them early, but they came out too. After taking the pic I realised that I'd placed some of the wrong values down. They are all in the right position, but just not necessarily of the right value... Mr. Preview.
Anyway, there's still the phono amp input stage which is yet to have its resistors changed as it's noiser than it should be.
Oh and many thanks to the previous poster whose hand-drawn circuit I used.

Gain for the output stage turned out to be a bit under 100, for a 20mV RMS input, I got a shade under 2V RMS out..... at all frequencies. So thanks for help.
David
Attached Thumbnails
Click image for larger version

Name:	repaired2.jpg
Views:	99
Size:	114.9 KB
ID:	288548  
Impecunious is offline   Reply With Quote
Old 23rd Nov 2023, 1:32 am   #16
Herald1360
Dekatron
 
Herald1360's Avatar
 
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,511
Default Re: Truvox TSA 100 Amp and OC81Z

Quote:
Originally Posted by Impecunious View Post
HOO ****** Ray
Sorted, at last and thanks for all your help.

.....


Oh and many thanks to the previous poster whose hand-drawn circuit I used.


.....


David

That's a blast from the past

https://www.vintage-radio.net/forum/...29&postcount=6

A pink pocketfile folder comes to mind; where it is now though?
__________________
....__________
....|____||__|__\_____
.=.| _---\__|__|_---_|.
.........O..Chris....O
Herald1360 is offline   Reply With Quote
Old 26th Dec 2023, 8:05 pm   #17
steve ss
Tetrode
 
steve ss's Avatar
 
Join Date: Mar 2023
Location: Victoria, Australia.
Posts: 55
Default Re: Truvox TSA 100 Amp and OC81Z

This is an interesting read!

A few points :

I assume the picture you have posted is of your amplifier and if so
it appears that it is in fact a TSA200 not TSA100 as all the front end transistors
appear to be silicon not germanium as in the TSA100.

The OC81Z is a higher voltage version (Vce) of the OC81D probably hand selected!
The D in OC81D denotes a driver transistor to be used with a flag heat sink.

As it happens I have a number of TSA100's and one had a failed OC81Z and I
replaced it with a very late miniwatt AC128 and so far (the last 15 years) it has
been working fine.

I noticed you used an OC84D which has a Vce =20 volts similar to the AC128.

You also mentioned that a number of resistors were noisy and this is typical of
a unit that has done a lot of "ON hours" which also indicates the electros as well
as most if not all resistors need replacement. The early polyester caps could be up graded to advantage also!

It should be possible down the track to find the correct switch pot and evict the
toggle switch.

A nice TSA100 Review.

Link:

https://rubli.net/classic_amps/files...vox/tsa100.htm

I hope this adds to the joy of ownership!

Enjoy Steve ss.
__________________
Contact your customer. Understand their needs. Then flog them something they could well do without.
steve ss is offline   Reply With Quote
Old 26th Dec 2023, 9:43 pm   #18
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,686
Default Re: Truvox TSA 100 Amp and OC81Z

Transistors are pretty non-linear and there are large variations in gain from unit to unit.

Consequently, audio amplifiers using semiconductors must be ruled by feedback. Strong feedback, lots of feedback. If the gain of a stage is being reduced by something wrong within the loop, then there will most usually be large distortion through clipping as something runs out of steam.

If your gain is low and you aren't getting noticeable distortion, then something is wrong either with the ratios of the feedback resistors, setting the amplifier the wrong gain as its goal, or else it is something outside of the power amplifier.

If the sound is still OK (but too quiet) across a range of volume, this reinforces the steer away from suspecting the main power amplifier.

Transistor amplifiers have serious imperfections and without utterly dominant feedback would be rather nasty. With it, they can be made very satisfactory. All the feedback in the world won't make an amplifier swing any further if something is wrong and is imposing a hard limitation.

So the first test with just your lug holes can give you a good guide as to whether you're heading into the main part of the amplifier or into the feedback resistors at most... or the preamp.

The next thing would be to use a meter and check that the mid-point of the output transistors is in the general vicinity of half the supply voltage for single supply amps. If it's a later style with positive and negative supplies to allow DC coupling on the output, then the mid point should be close to zero volts. There may be an adjustment to optimise this.

After this check the bias current. Output stages usually amount to emitter followers. In this case the lower transistors aren't a Darlington in emitter-follower configuration, it's a compound transistor simulating an NPN power device. However, a voltmeter across the low value emitter resistor and Ohm's law will get a current reading. Again, there is likely to be a twiddler to set this to the stated design value. There can be some interaction, so check the mid point voltage again.

Getting this far gives some confidence that the bulk of the power amplifier is OK.

The next job is to measure the feedback resistors. If one has gone high or low, it will change the gain value that the feedback controls the overall power amp gain to. Transistor amplifiers are a lot on the wild side. Without feedback to tame them they will exhibit ludicrously high gain and will probably try to ram themselves against one end-stop or the other. Device variations allied with the high gin pretty much ensures voltages will be slewed close to one limit or the other unless there is a DC controlling loop working.

This amplifier has THREE feedback loops at work.

There is a DC midpoint negative feedback path via R40

There is an AC POSITIVE feedback path via C23. This deserves some explanetion. It allows the output swing to drive R39, so that R43 sees matching AC components at both ends and simulates a constant-current source allowing the gain of VT6 to be much higher than you'd guess from the resistor values alone. It also allows the drive to the emitter follower output stages to swing right to the rails and even a little beyond if needed. This is normally referred to as 'Bootstrapping' but it's a bit clearer if it's thought of as unity gain positive feedback and the simulation of close to infinite impedance.

Thirdly, this amplifier has its main AC feedback to an earlier AC amplifier stage VT5 This is the feedback path which sets the overall AC gain for signals, which reduces distortion, reduces the output impedance, mends split ends and prolongs active life.

I hope this run-through helps a bit.

VT6 is in a position usually called the VAS, Voltage amplifier stage. It does the bulk of the voltage gain of the amplifier. The four transistors following it are principally current gain stages of unity voltage gain. They do all the heavy lifting.

The VAS in amplifiers places pretty tough demands on the transistor used. Remember that bootstrap path can drive the collector voltage beyond the rails. The bootstrap capacitor will faithfully transport any nasty transients and surges from switching at the output end, straight back to the collector of the VAS. VAS transistors are very prone to failure, often due to second-breakdown mechanisms. The headphone jack is just right for helping make transients as it switches speakers out.

It is possible that the AF118 has gone a bit leaky and to an extent the DC loops will compensate. You'll mostly see increased temperature though the amp may well soldier on. Amplifiers so heavily dominated by feedback can cover up all sorts of minor faults or damage and sound to be working normally. You may not be able to tell by looking at the output. To decide whether an pmplifier is really OK and that there isn't a bigger problem coming down the line, it's necessary to dig in and actively look for trouble.

A final check would be to drive the amplifier from a signal generator, using a realistic load on the output and watching the output waveform on an oscilloscope. Wind up the signal level until you see the beginning of clipping at one extreme of the sine or the other. Measure the voltage swing and calculate the output power. If there is something wrong being covered for by the surplus gain and feedback, it will show up here.

If you're used to valve amplifiers, transistor ones have a lot of culture shock. They're not bad, not evil, just different. People look at the circuitry and compare it to familiar valve circuitry, but this misses the huge difference in feedback design. Without transformers in the way, a major source of high frequency phase lag is missing. Also impedances in earlier stages are a lot lower, also leading to wider bandwidth and less phase lag. Also transistors are small, cheap and don;'t need heater power. It's easier to employ more of them. It took time for amplifier designers to stop trying as hard as they did with valves to minimise the number of active devices.

This improved bandwidth and phase response allows a lot more feedback to be deployed without running into stability problems. That's good. Transistors need a lot of feedback.

Somehow, it all fits together and really good amplifiers can now be made with transistors. There was a dodgy patch while designers were learning. This and the Leak stereo 30 etc. are from the learning period, which makes them interesting. It was a period when things were advancing at a fast rate.

David

Afterthough: If the top AD140 is down on gain, then the OC81Z will be having to carry a heavier proportion of the load current and bias current, which will get it hotter than it should be.
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now   Reply With Quote
Old 8th Jan 2024, 1:29 am   #19
Impecunious
Triode
 
Join Date: Oct 2011
Location: Birmingham, West Midlands, UK.
Posts: 28
Default Re: Truvox TSA 100 Amp and OC81Z

David, Radio Wrangler,
Many thanks for the detailed explanation as to how the amp works, much appreciation for taking the time and effort to explain clearly.

In the end it was a feedback problem, VR52 had gone high, causing that channel to have excessive gain. For ages I had accused (and spent a long time fault finding within it ) the other channel of being the miscreant, simply because its gain was lower than the other channels gain, when all along it had been working perfectly.
D
Impecunious is offline   Reply With Quote
Old 8th Jan 2024, 1:40 am   #20
Ted Kendall
Dekatron
 
Join Date: Jul 2008
Location: Kington, Herefordshire, UK.
Posts: 3,644
Default Re: Truvox TSA 100 Amp and OC81Z

I did that too, years ago. One channel of my Texan was apparently down compared to the other, which had so much gain the signal forced its way past a closed volume control. It was a while before I realised there was a dry joint in the feedback around an op-amp in the louder channel...
Ted Kendall is offline   Reply With Quote
Reply

Thread Tools



All times are GMT +1. The time now is 8:11 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.