I am sharing one publicly available document by Texas Instruments discussing exactly this topic of Hall Effect based vs Shunt based sensors used in Electric Vehicles. This is a recent document 2018-2019.
Quoting some important points from the article.
Historically, designers preferred shunt-based solutions for low-current (<50 A), and hall-based solutions for high-current (>50 A) measurements.
However, because of the increasing current measurement accuracy requirements, automotive suppliers are migrating from hall-based to shunt-based methods, especially in high-current environments.
Multiple measurement methods exist for isolated current sensing in HEV/EV subsystems including shunt-based and hall-based methods. With advancements in affordable high-precision shunts and high-performance isolated amplifiers and modulators, shunt-based solutions have become good alternatives to traditional hall-based solutions.
From another published paper (much earlier maybe 2007-08):
A standard current-measuring transducer with average performance characteristics is able to provide very high-level of accuracy when used in current-integration applications. This high level of performance can be achieved because magnetic offset and gain errors are cancelled when a transducer with good linearity is used with a fairly well balanced current profile cycle that is typical in HEV or internal combustion vehicle applications with battery management systems. The realization of this behavior allows the focus of attention to be on the quality of the measurement results, and not on the benchmark performance characteristics of the transducer being considered. Furthermore, it allows an optimized-cost solution to be determined quickly by understanding the real-world performance of a particular transducer, instead of it’s laboratory benchmark values. This is why for the past several years now, the world’s top car manufacturers have been using open-loop Hall-effect transducers in their battery management systems.
But another article quotes that earlier cars such as Toyota Prius (an HEV) were primarily using HE based sensors but there is a trend towards Shunt based ones on the lines of the Texas instrument article.
My conclusion as of now is that in the EV applications there is a trend towards Shunts. Please correct me if I am wrong. But I dont know why Shivalik is the only producer. I havent studied the company in detail.
Disclaimer: I am generally reading these research papers on EVs. I am no expert.
The obvious question that comes to mind is how is Shivalik the only producer of Shunts. Is there any competitive edge as to why other players are not jumping in. I understand that Shivalik must be having existing client base, but given the whole noise around EVs, most manufacturing companies would be looking to target every small part which could be used in an EV.
For lower rated current we can use shunts. For higher rated current one needs HE sensors. 2W & 3W are low powered batteries around 5 to 6kwhr. For 4 wheel one requires a min 15kwhr battery. So high current will flow and Shunts are not the best alternative.
Will require more views from the experts to establish this.
Shunt based current sensors are new where as Hall effect sensors are traditional way of measuring current which is then converted to voltage with additional electronic circuits. Hall effect sensors has offset, i.e. it doesn’t measure zero when there is no current and this error increases with time where as shunts have no such error. Another positive for shunts is its operation on high temperature. Shunts can easily support 150C where as Hall effect sensors are useful only up to 85C. This is huge advantage as 85C is non automotive grade and automotive needs electronic components to work 125C MINIMUM. One advantage Hall effects sensors have is isolation. As batteries are grounded to chassis so if there is no isolation between sensor and battery terminal we would get shock. Shunts need additional circuitry for this isolation where as Hall effect sensors are by design do not touch battery terminals and work on magnetic field effect so no additional isolation circuitry cost is needed. However Hall effect sensors need additional circuitry to deal with offset described above. As on today until these limitations of Hall effect sensors are dealt with we can expect shunts to move ahead in adoption. 85C electronics is strict no to have a component qualified for auto grade reliability and this is one big limitation. The accuracy is non linear above this temperature in Hall effect sensors and shunts are quite stable with up to 1% temperature tolerance. Until this limitation is overcome in Hall effect sensors shunts has an edge.
Shunt manufacturing requires material science and electronics knowledge that Shivalik has and they have started at right time. We had passive components shortage in 2017-18 when EVs suddenly started absorbing all passive components capacity world over suppliers worldwide started looking at alternative sources. I think Shivalik was ready with required products so got big boys of the domain as customers. Automotive is all about reliability. Automotive is usually one generation behind in latest technology available. Say our laptops already had DDR4 but you would not find it in Automotive yet. Reason is automotive reliability testing. The automotive technology providers are struggling to qualify DDR4 to auto grade and we not only are finding new issues but even test equipment’s used needs tweaking. Shivalik is IATF 16949 certified and it takes lot of efforts to build this capability. If Shivalik shunts are auto qualified by a car OEM they would not change their vendor to a new vendor or until they qualify their solution with new vendor again. This works as a moat once we have one component in a car. Auto qualification takes 18 months minimum. Globally shivalik would need to engage with global automotive leaders or OEMs to play at global level. They are with leader so thats good start but they would need to get it qualified and become an approved supplier. Once they are approved with some big name like continental others would come automatically.
I think key monitorable at the moment is if Shivalik is an approved vendor with a global automotive vendor? We shall also monitor if Hall effect sensors have dealt with their temperature issue and can it be made available in similar cost.
In BMS we need to monitor each and every cell that contributes to overall wattage of the battery. Each cell would draw much less current than overall battery. Heard Tesla uses shunt registers to monitor its cells. There are shunt based BMS designs that can sustain
Please note that sensing is not the only place where SHUNTs are used. We need them in protection circuits as well.
Company seems to have moat in the product and business but promoters are juicing it even when the business was in trouble ( period when CRT products became obsolete)
@kalpesh4430 - yes, management remuneration has been on the higher side and this year the company raised the salary further despite company not doing well in the current year and this had upset most of the shareholders at AGM. I don’t know why management’s don’t understand simple maths and work towards value creation which will benefit the promoters the most.
@rambaranwal - thanks for your inputs. Please share more insights as you come across.
BMS market is clearly dominated by the 3 German Tier 1s. They have a huge pioneer advantage as they started as early as 2014s. The rest are followers
Low ohmic shunt looks to be dominated by Isabellenhuette with over 50% market share
Murata has an established chain of Tier 3s from Rohm and Manudax
Who is supplying to Vishay as Tier 3 I was not able to find (still back of our global purchasing experts awaiting his reply)
Shivalik was never heard at Marquardt, not even engineering samples are present in the pre-development (it is a standard practice to test either low-cost or new technologies years before they are implemented in the products). I will try further to check with my contacts in Conti Germany what is the situation with them, too many holidays right there.
Inventory glut, however, seems to be the case as 2019 they have built up the stocks of most of the passive components, which came as the boon in disguise for 2020.
@rambaranwal The no. of shunts used clearly depends on the safety requirements defined by the OEM. What you said is right in the initial generations of BMS where every cell used to be measured. Then it was per pack (a combination of cells) and in today’s case, it is 1 shunt per BMS or max 2. It still might be the case if a particular OEM says to the BMS supplier to measure per cell. But that is mostly not the case I understood, as we are in 5 or 6th generation at least with Conti and Marquardt
@rambaranwal or any of you - if we have any contacts in Conti India Engineering centre? Hardware guys working on these BMSes can help us further understand the routes of supply. Checking that option too.
@Vpayasam
Bravo! This is Super value-addition! at Shivalik thread after a long long time
Auto domain veterans within VP like Varun Singla please wake up. And connect us to some Senior Marketing/Supply-Chain guys at Continental or Hella. Think the Shivalik puzzle is now solvable in quick time. All Auto Domain professionals - please update profiles, asap - Your urgent help is needed to connect to right folks
a) Inventory Glut situation - the part we need to focus ON/ get more CLARITY, quickly.
b) Tier2 supply is more likely supplying complete PCB mounted with Shunts and other Passives (resistors, semiconductors)?? (not Shunts but complete assembly of passives for BMS)
c) Tier3 - While Isabellen Hutte is the largest producer of Shunts, Vishay we know sources from Shivalik (both strips and shunts?). So we have 4 established names now as bulk producers of shunts - Vishay, Isabellen Hutte, Rohm Semiconductor, Shvalik.
Who are the other two - Chinese players?
We should next establish Vishay and Rohm approx Mkt Share for low-ohmic shunts
that would leave us with addressable market share for Shivalik as probably like 10-15% of global market (assuming the 2 others are shivalik scale)
d) Or maybe Shivalik, and others like it are more like Tier 4?
Think we should prioritise access to Denso?
Likely to get more nuanced/granular competitive info from there. The idea will be to map out the scope for multiple commercil relationships from Shivalik - besides Vishay. One the face of it (above picture if completely true) looks grim - supply relationships are firmly entrenched - everyone has stocks. Breaking into a relationship will happen - only there is a shortage situation again? How far out is that?
Many subsuppliers to tier 1 – List is endless and shivalik is one of them. Industry growth rate 6.4% until 2025 as per supplier TTE
Shunt resistor seems to be superior technology to Hall effect sensor
Not only BMS, many more usage in automotive and other battery related products(scooter, Vacuum cleaner).
Seems Other applications provide more growth opportunities than BMS
Points related to Shivalik:
Shivalik is subsupplier to Vishay and almost 20 Tonnes of delivery to Vishay
Vishay is major competitor with large product base and quality as per competitor TTE and shivalik seems to be one of the main supplier based on shipping order
Shivalik seems to supply to 3 japanese companies and one of them is japan Resistor. Bimetal and Seidentechno are other 2 customers. Their product line up matches with shivalik product range.
As per company Bimetal, shivalik seems to have quality product which fulfills japanese requirements.
Further Work:
Discussion with industry technical guys are lined up to understand the usage and quantity pattern.
Great to see you delivering quietly with solid value-addition, again. If we keep this up, very soon VP will have much more fine-grained and nuanced understanding of BMS for EV Market. @Vpayasam@rajanprabu
Prakash too is based in Germany. You guys should collaborate with each other, discuss and crack open this puzzle soon.
Now few observations:
a) we are primarily interested in BMS for EV Market (different specs, different price-points, different competitive map for other applications? good to know them though)
b) lets filter out the Competitive Map to just BMS for EV. Prima facile looks like a few Japanese names get added to Tier2/3
c) keep the FOCUS firmly on Market Share mapping - I all our scuttlebutts - to zero in on the biggies
d) keep the FOCUS firmly on getting down to Inventory glut sizing, recovery timeframes
e) Scuttlebutt PDF provided is very good and in the right direction - thank you; good to indicate Sources (even if not public) for cross-checking by others
c) and d) probably move the needle the most
Be aware that you often get seemingly contradictory/conflicting information most times from different scuttlebutt sources.
As we keep talking to folks in the domain, we will learn from each interaction and will be able to use that incremental learning productively with the next expert … you get the drift…and finally arrive at the key(s) to the puzzle.
What more can be done? What are we missing? Time for domain experts to rise to the occasion and guide us better.
We are missing the automotive domain experts within (and outside of) VP, for sure
@wandering - haven’t seen such hard and detailed work. Nice connecting of dots! You have been researching on this company from before or did it recently?
Overall what would be your thoughts on 1. key things to monitor going forward 2. key risks 3. what is opportunity/potential here?
BMS market growth - Impact of COVID on Auto Industry - Recovery trajectory
Shivalik breaking contract with Continental or Hella for Tier 2 supply (first we need clear proof that both of them are working towards it). This would drive the margins higher and customer trust will increase significantly and aid in gaining new customers (BMS is closely consisted market with major players being from Germany - also they do deliver to non EV Market - For e.g. Marquardt has a separate Business unit concentrating on Pedelecs, E-Bikes and Consumer Products )
Inventory position at Vishay and OEMs where Shivalik is delivering
Risks:
Average no. of shunts being 1-2 per BMS
Shivalik not able to match the measurement precision of “Isabellenhuette” despite the cost advantage
Opportunity/Potential:
BMS market has significant growth potential - given the EV penetration worldwide - Considering 2022 estimates 7.25 B $ with an Avg BMS cost of 60$ and BOM cost of 1 $ for shunts (based on average quantity- conservatively as consumer products would need more) market size we are talking of 120 M$ i.e. 906 Crores. Hence, one wouldn’t be wrong in saying the market and the potential is significant for Shivalik
@wandering - Prakash sent you a PM lets catch up on this offline to dig further!
In case someone missed to notice the following good gesture from the promoters - gives some great signal on management honesty:
Pursuant to Regulation 30 of SEBI (Listing Obligations and Disclosure Requirements) Regulations, 2015, as amended, please take note that on account of the COVID-19 scenario globally which has impacted the Indian as well as Global economy and current lock-down within the country, Mr. S.S. Sandhu, Chairman and Mr. N.S. Ghumman, Managing Director have decided to repudiate their salary for 3 months starting from April 1, 2020 till June 30, 2020, in order to strengthen the liquidity position within the Company.
Though it appears to be a good gesture, its difficult to conclude anything from this. There are lot of examples where owner promoters company CEO’s waving off salaries or taking very small amount.
Lets not conclude anything out of this. It just adds some feel good factor while someone is evaluating the management quality. They have a history of standing by the company during the tough times, however taking the money out when the profitabilty is back.
When the company was going through a rough patch, directors had foregone 70 per cent of their entitlement, amounting to Rs 62.64 lakh in director remuneration and opted to avail only 30 per cent of the increased remuneration as approved the central government for the FY 2011-12.
Further, in FY14, due to constrained profitability, the management voluntarily proposed to work on the same salary for next three year.
However, in FY18, when the company resumed on profitable path, salaries of promoter directors increased to Rs 3.63 crore against a net profit of Rs 15.99 crore, which was higher than total dividend payment of Rs 2.77 crore during the year.
Referred following article (details can be found in ARs):
Concerns already raised and discussed in this thread regarding remuneration on the higher side and that is still a concern. However, repudiating salary during this crisis also give some comfort level to that concern.
New Scheme in India for Electronic Manufacturing opens up to 1st Aug and Govt expects at least 5 Global and 5 Local Electronic Manufacturers to join the scheme. This also includes component manufacturing. If this takes off positive for companies like Shivalik
soon.
